Content uploaded by Alison Perrin
Author content
All content in this area was uploaded by Alison Perrin on Nov 25, 2015
Content may be subject to copyright.
Economic Implications of Climate Change
Adaptations for Mine Access Roads in Northern Canada
This publicaon may be obtained from:
Northern Climate ExChange
Yukon Research Centre, Yukon College
520 College Drive
P.O. Box 2799
Whitehorse, Yukon
Y1A 5K4
867.668.8895
1.800.661.0504
www.yukoncollege.yk.ca/research
Recommended citaon:
Perrin, A., J. Dion, S. Eng, D. Sawyer, J.R. Nodelman, N. Comer, H. Auld, E. Sparling, M. Harris, J.Y.H.
Nodelman, L. Kinnear. 2015. Economic Implicaons of Climate Change Adaptaons for Mine Access
Roads in Northern Canada. Northern Climate ExChange, Yukon Research Centre, Yukon College, 93 pp.
Photo credit: Dave Brosha
With funding from Natural Resources Canada in
support of Canada’s Adaptaon Plaorm.
ACKNOWLEDGMENTS
The project team would like to thank all the participants for their enthusiasm and commitment.
We express our appreciation to Risk Sciences International, International Institute for
Sustainable Development, Yukon Research Centre, the Tibbitt to Contwoyto Winter Road Joint
Venture, NOR-EX Ice Engineering, Nuna Logistics, Tetra Tech EBA, Inc., the Government of
Northwest Territories and all those noted in the following pages for their support. We especially
thank the participants who attended the workshop in Yellowknife, Northwest Territories and
provided feedback and technical advice throughout the project.
Funding for this project was provided by Natural Resources Canada, Government of Canada in
support of Canada’s Adaptation Platform Northern Working Group. Project management was
conducted by the Northern Climate ExChange, part of the Yukon Research Centre at Yukon
College in Whitehorse.
PROJECT TEAM
Lead authors
Alison Perrin Northern Climate ExChange, Yukon Research Centre, Yukon College
Jason Dion
International Institute for Sustainable Development
Simon Eng Risk Sciences International
Dave Sawyer EnviroEconomics
Joel R. Nodelman Nodelcorp Consulting, Inc.
Contributing authors
Dr. Neil Comer Risk Sciences International
Dr. Heather Auld Risk Sciences International
Erik Sparling Risk Sciences International
Melissa Harris
International Institute for Sustainable Development
Joan Y.H. Nodelman
Nodelcorp Consulting, Inc.
Lacia Kinnear Northern Climate ExChange, Yukon Research Centre, Yukon College
Technical Advisors
Pietro de Bastiani Government of the Northwest Territories, Department of
Transportation
Al Fitzgerald NOR-EX Ice Engineering Inc.
Matt Fournier Government of the Northwest Territories, Department of
Transportation
Don Hayley
Geotechnical engineer
Tom Hoefer Northwest Territories and Nunavut Chamber of Mines
Ed Hoeve Tetra Tech EBA, Inc.
David Lapp Engineers Canada
Dane Mason Government of the Northwest Territories, Department of
Transportation
Kendra McGreish Government of the Northwest Territories, Environment and Natural
Resources
Ron Near
Tibbitt to Contwoyto Winter Road Joint Venture Management
Committee
Katrina Nokleby NOR-EX Ice Engineering Inc.
Brian Sieben Government of the Northwest Territories, Environment and Natural
Resources
Jim Sparling
Government of the Northwest Territories, Environment and Natural
Resources
Jennifer Stirling Tetra Tech EBA, Inc.
Tim Tattrie Nuna Logistics
Rob Thom Government of the Northwest Territories, Department of
Transportation
Robert Zschuppe Tetra Tech EBA, Inc.
Technical Editing and Production
Patricia Halladay Editor, Whitehorse
Guin Lalena Graphic designer, Whitehorse
EXECUTIVE SUMMARY
Climate change is one of the major threats to northern infrastructure in Canada. The mining
industry is particularly vulnerable to climate change, and in the Canadian North the effects of
climate change on mining-related transportation have become a significant concern. Climate-
driven disruptions to mine access roads have caused economic losses and raised concerns about
projected future changes in the region.
This study evaluates the climate-related vulnerabilities and related costs and benefits of the
Tibbitt to Contwoyto Winter Road (TCWR), a mine access road built mainly over frozen lakes in
the northeastern region of the Northwest Territories (NWT). The TCWR is the main access road
for three active diamond mines, as well as one that will be active in the next year, and is the
busiest heavy-haul ice road in the world. Diamond mining in this region has been a major
contributor to the NWT economy. The road is operated by a joint venture between three mining
companies, with contracted support for engineering, maintenance and security. Companies
transport goods via the winter road to supply year-round operations. When the road season is
cut short, supplies must be transported by plane or helicopter, which greatly increases costs.
One of the main goods being transported to the mines is diesel to power mine operations.
The vulnerability study identified five climate variables that affect the viability of the TCWR to
supply the mines:
• operational season length (interaction of freezing-degree days and melting-degree
days);
• incidence of temperature swings in excess of 18˚C;
• incidence of consecutive days above 0˚C during the operational season;
• amount of snow on the ground January 1; and
• number of extreme cold events during the operational season.
Under historical climate conditions, the TCWR is generally resilient; however, from the analysis
of these five climate variables, the length of the operational season — driven by temperature —
was identified as the key vulnerability. Two future scenarios were identified for the road based
on the vulnerability of the road’s operational season length to temperature (freezing-degree-
days and melting-degree-days) and an analysis of potential adaptations.
The two scenarios capture the impact on the road of increasingly difficult climate conditions.
The first is an adaptation scenario based on difficult climate conditions, leading to shorter
operational seasons as maintenance and repairs become more difficult and costly. The second is
a critical conditions scenario based on highly challenging climate conditions, leading to late
opening, early closure, or non-opening of the road where the desired levels of road access
become impossible and other modes of transportation are required to move loads.
The economic analysis evaluated the impact of multiple climate variables and their thresholds
on the two scenarios. The affected cost types, affected stakeholders and thresholds were
identified for each variable.
The key cost types that increased net costs for the adaptive scenario include flexible scheduling,
and increased construction and maintenance for the ice road, portages and ramps. The greatest
cost increase is from adaptive scheduling on the part of carriers. If the season is shortened
below 50 days, additional costs are triggered as loads are shifted. Over the assumed 35-year
time horizon the total estimated costs of the adaptive scenario are in the order of $55 million.
This was the mean value produced by the economic analysis, with a maximum cost value of
$155 million and a 60% probability that the actual value of the costs would be greater than $55
million.
An operational season length of less than 45 days was identified as the trigger for the critical
conditions scenario. At that point the road would no longer be able to accommodate an average
season’s demand. The two key cost types that increased net costs for the critical conditions
scenario are using alternative forms of transportation and production loss. The expected total
cost over the 35-year time horizon is $213 million, of which production loss is the dominant
cost, at approximately 70% of the total. The maximum value for this scenario is $1.8 billion, with
a 60% probability that the actual value of the costs would be greater than $213 million. The
results of the analysis are highly sensitive to the assumed forecast in operational season length,
which implies that small changes in season length would result in large and significant future
damages under this scenario.
Operational season length is the most important cost driver in the future, partly because some
of the other climate-cost variable interactions are likely to become less of a concern with
climate change, or remain unaffected by it. Some of the less important climate-cost
interactions, such as construction costs, have much smaller incremental effects even when
triggered by the adaptive scenario. The operational season, on the other hand, has a significant
impact on carrier costs in the adaptive scenario and on both the need for alternative
transportation and production costs in the critical conditions scenario. The results indicate that
if the demand continues at current levels, the expected evolution of operational season length
creates a significant economic risk to TCWR operators and users.
TABLE OF CONTENTS
1.0 INTRODUCTION ........................................................................ 1
2.0 CASE STUDY .............................................................................. 2
2.1 ROAD DESCRIPTION ................................................................................... 3
2.2 GEOGRAPHICAL ELEMENTS ....................................................................... 4
2.3 ROAD DESIGN AND CONSTRUCTION ......................................................... 5
2.4 ROAD TRAVEL ............................................................................................ 6
3.0 ESTIMATING ECONOMIC IMPLICATIONS ................................... 7
3.1 STEP 1: HAZARD, VULNERABILITY AND ASSETS-AT-RISK ........................... 8
3.1.1 Screening Assessment ............................................................................................ 8
3.1.2 PIEVC Hazard Assessment .................................................................................... 10
3.1.3 Adaptation Responses .......................................................................................... 15
3.2 STEP 2: CLIMATE CONDITIONS AND FORECAST ...................................... 17
3.2.1 Projected Climate Variables ................................................................................. 19
3.3 STEP 3: ASSESSMENT SCENARIOS ............................................................ 22
3.4 STEP 4: ECONOMIC ASSETS-AT-RISK AND FORECASTS ............................ 23
3.5 STEP 5: CLIMATE RISK AND ECONOMIC VALUE ASSESSMENT ................ 30
3.6 STEP 6: NET COSTS OF CLIMATE-INDUCED WEATHER PATTERNS .......... 34
4.0 RESULTS .................................................................................. 34
5.0 CONCLUSIONS ........................................................................ 37
REFERENCES ......................................................................................... 39
APPENDIX A: HISTORICAL CLIMATE ANALYSIS ...................................... 43
A. 1 INTRODUCTION ........................................................................................ 44
A.2 HISTORICAL CLIMATE FACTORS ............................................................... 44
A.2.1 Winter Temperature ............................................................................................. 45
A.2.2 Freezing-Degree Days (FDD) and CANGRD temperature values .......................... 49
A.2.3 Melting-Degree Days (MDD) ................................................................................ 54
A.2.4 Date when Accumulated FDD reaches 300 .......................................................... 56
A.2.5 Accumulated Snowfall as of January 1 ................................................................. 57
A.2.6 Snow on Ground (measured) as of January 1 ...................................................... 58
A.2.7 Accumulated Rainfall in November and December ............................................. 59
A.2.8 Days with 24-Hour Temperature Change Greater than 18°C ............................... 60
A.2.9 Days with Mean Temperature over Freezing ....................................................... 61
A.2.10 Days with 24-Hour Temperature Drop Greater than 20°C ................................. 62
A.2.11 Observed Lake Ice Measurements (Yellowknife, Great Slave Lake)................... 63
A.3 GORDON LAKE LOCATION ....................................................................... 64
A.4 SUMMARY OF CLIMATE FACTORS INFLUENCING THE TCWR ................. 67
A. 5 CONCLUSIONS AND FURTHER WORK ...................................................... 71
APPENDIX B: FUTURE CLIMATE ANALYSIS ............................................ 72
B.1 INTRODUCTION ........................................................................................ 73
B.2 PROJECTION DATA ................................................................................... 73
B.3 PROJECTION METHODOLOGY.................................................................. 79
B.4 PROJECTED WINTER TEMPERATURE ....................................................... 80
B.5 PROJECTED WINTER FREEZING-DEGREE DAYS (FDD) .............................. 82
B.6 PROJECTED FDD AND WINTER ROAD OPERATING PERIOD ..................... 84
B.7 PROJECTED DATE OF ACCUMULATED FDD AT 300 ................................. 86
B.8 PROJECTED MELTING-DEGREE DAYS (MDD) ........................................... 88
B.9 CONCLUSIONS .......................................................................................... 90
REFERENCES: APPENDIX A AND B ......................................................... 92
FINAL REPORT
Economic Implications of Climate Change Adaptations for Mine Access Roads
in Northern Canada
1
1.0 INTRODUCTION
Climate change has widespread implications for both private and public infrastructure in
Canada’s North. The mining sector is a fundamental component of the Canadian economy,
contributing more than $10 billion per year and employing almost 50,000 people in primary
mineral extraction alone (Pearce et al. 2009). The northern mining sector makes up 20% of
Canada’s mineral production (Pearce et al. 2009).
The mining industry’s dependence on the natural environment makes it particularly vulnerable
to climate change. In the Canadian North, the effects of climate change — particularly on
mining-related transportation — have become a significant concern (Perrin et al. 2015). Over
the past decade, some mines have experienced unusually significant economic losses as a result
of climate-driven disruptions to mine access roads, including shortened winter road seasons
and large road washouts due to flooding (Ashbury 2006). Projected increases in extreme
weather events and changes in average climate conditions will likely disrupt ground-based
transportation routes and operations even further, through increasingly significant effects on
permafrost, lake and river ice, higher numbers and intensities of land and snow slides, severe
runoff and flooding events, and erosion, among other factors.
Northern jurisdictions have identified the need for greater understanding of the implications of
climate change on infrastructure, and of how to quantify the costs of adapting to this change.
Key policy documents such as Canada’s Northern Strategy (INAC 2009) and the Pan-Territorial
Adaptation Strategy (GNWT, GN and YG 2011) emphasize the importance of critical northern
industries working to further assess and manage the risks of climate change.
Generally speaking, ground-based access has been extremely important to the production,
safety and economic growth of the northern mining sector (Nelson and Schuchard 2011; Pearce
et al. 2009). As a result, northern governments, professional associations and the mining sector
have all expressed interest in enhancing their understanding of options for adapting to the
impacts of climate change. Proactive adaptation to climate change can lower the overall cost of
climate impacts by moderating or preventing damage, shortening service disruptions, and
reducing risks to human health and safety (NRTEE 2011). Relatively little work has been done in
northern Canada to assess the costs and benefits of adaptation with respect to mine access
roads.
This project brings together the expertise of the Northern Climate ExChange, the International
Institute for Sustainable Development, Risk Sciences International, Nodelcorp Consulting and
EnviroEconomics to develop a cost-benefit analysis of a range of adaptation options for a major
northern mine access road. The project focuses on a case study analysis of the Tibbitt to
YUKON RESEARCH CENTRE
Case Study of the Tibbitt to Contwoyto Winter Road
2
Contwoyto Winter Road (TCWR). The road was chosen due to its sensitivity to climate and its
regional economic importance, and because of the availability of existing research and analysis.
Mining is a key part of the economy of the Northwest Territories, and the viability of the TCWR
in the future is a key concern for the existing and potential mines along its route.
This report provides background information on the road, including projected climate
conditions for key parameters related to road operations, a step-by-step description of the
methodology used to estimate the economic implications, a summary of the results of that
analysis, and some key findings from the project.
The results of this study provide valuable information for road owners and stakeholders using
the Tibbitt to Contwoyto Winter Road, and for decision-makers planning for changes to the
road. The results provide a better understanding of climate thresholds for future years and how
crossing those thresholds will affect costs. The economic information provided in Sections 3.4 to
3.6 will help road owners anticipate the magnitude of costs that they may incur during shorter
seasons. With shorter seasons potentially happening more frequently, the findings provide
additional information for decision-makers to consider as they assess whether to build an all-
season road.
The findings can also inform future planning, economic assessments and vulnerability studies
for other northern roads, particularly winter roads. The methodology can be applied to other
supply roads in northern jurisdictions, and the information on key climate vulnerabilities can
inform other winter road managers.
2.0 CASE STUDY
The Tibbitt to Contwoyto Winter Road (TCWR), formerly known as Echo Bay Mines Limited’s
Lupin Winter Road, was originally constructed to supply the Lupin Gold Mine at Contwoyto Lake
in what is now Nunavut. The road currently supplies four mines and is the only overland supply
route. It provides access for reclamation efforts at former mines, including the Tundra Mine
(which was reached by a spur road during the 2015 season).1 It also supplies some exploration
properties, contaminated site remediation projects, and tourism and outfitting camps, although
third-party traffic of this type is minimal (Greenspan 2008).
The road is operated by the Tibbitt to Contwoyto Winter Road Joint Venture Management
Committee (JVMC), which is made up of the Dominion Diamond Ekati Corporation (Ekati mine),
Diavik Diamond Mines Inc. (Diavik mine) and DeBeers Canada Inc. (Snap Lake and Gahcho Kué
1 A winter road year is the year in which the operating season occurs (i.e., January – April)
FINAL REPORT
Economic Implications of Climate Change Adaptations for Mine Access Roads
in Northern Canada
3
mines). The JVMC is also responsible for contracting companies for construction, engineering
and security services for the winter road.
The TCWR is the busiest heavy-haul ice road in the world (Tetra Tech EBA 2013), with a record
10,922 loads and 330,002 tonnes (t) hauled in 2007 (JVMC 2013). Since 2001 the road has seen
increased traffic, which has also brought greater regulatory scrutiny (Tetra Tech EBA 2013). The
road is used to bring in diesel fuel, cement, tires and explosives supplies, in addition to other
manufacturing and construction materials and equipment.
The TCWR has been celebrated for advancing engineering practices for ice roads and for
contributing to the development of the northern mining industry. Notably, the JVMC won a
Professional Award of Merit from the Northwest Territories and Nunavut Association of
Professional Engineers and Geoscientists for “its combined engineering, human and socio-
economic achievements” (Tetra Tech EBA 2014).
2.1 ROAD DESCRIPTION
The road starts at Tibbitt Lake at the end of Highway 4, approximately 70 kilometres (km)
northeast of Yellowknife (Figure 1). Originally, the road was 600 km in length, ending at Lupin
Gold Mine at Contwoyto Lake, Nunavut, until it was extended to reach Shear Minerals’ Tahera
Diamond Corporation Mine (JVMC 2013). Neither of these mines is currently in operation and
the road now ends around the Diavik and Ekati Diamond Mines, at around the 400-km mark.
The TCWR travels generally north and northeast, and links to the Snap Lake Diamond Mine,
Diavik Diamond Mine and Ekati Diamond Mine, as well as the Gahcho Kué Project (Nuna
Logistics 2014). Camps, shops, laydown areas for equipment storage, and fuel stores are
strategically positioned along the road, which allows construction and maintenance teams to
respond quickly to issues as they arise.
The TCWR provides dedicated service to the operating mines along the route. The road must
deliver specified tonnages of goods to the mines over the operational season. If severe weather
events interrupt the operation of the road, or shorten its operating season, lost service can be
recovered in most cases by increasing the number of daily loads during the season. During the
operating season, road operations can be run later into the evening, or even 24 hours a day,
until the loss is recovered, the limiting factor being the number of trucks and drivers available.
This flexibility has a significant impact in reducing the severity of events. Even though road
operations are interrupted, the overall load levels through the operating season can be
maintained.
YUKON RESEARCH CENTRE
Case Study of the Tibbitt to Contwoyto Winter Road
4
Figure 1: Map of Tibbitt to Contwoyto Winter Road
Source: JVMC 2012
This is an unusual feature of this particular winter road. It may not be shared by other winter
roads in northern Canada, where there is more interest in using the road consistently over the
season as opposed to delivering a certain number of loads. For example, winter roads that
provide access to northern communities may not see the same amount of high-volume loads,
but will be used for as long as possible in the season.
2.2 GEOGRAPHICAL ELEMENTS
The TCWR travels north and northeast from Yellowknife through two ecozones: the Taiga Shield
and the Southern Arctic. It crosses lakes, streams, boreal forest, a transitional zone and
barrenlands terrain; most of the road passes over frozen lake surfaces. The overland traverses,
called portages, generally follow low-lying terrain, including frozen streams and wetland areas
near lakes. In hilly areas the road follows existing grades as much as possible. From Yellowknife
to the Mackay Lake region the road passes through the Taiga Shield Ecozone. This area is below
the tree line and the portages or overland stretches pass through the boreal forest. On some
portages, trees and taller shrubs have been cleared; other portages have been affected by
FINAL REPORT
Economic Implications of Climate Change Adaptations for Mine Access Roads
in Northern Canada
5
forest fires. North of Mackay Lake the road passes through the Southern Arctic Ecozone, which
is barrenland characterized by shrub tundra. The transition area between the two ecozones, the
Coppermine River Upland ecoregion, is made up of open stands of stunted trees (EBA 2001).
2.3 ROAD DESIGN AND CONSTRUCTION
Design and construction of the TCWR is a collaborative effort that is undertaken every year by
the JVMC, engineering consultants and the construction firm. The road is designed based on the
upcoming season‘s supply needs and schedules and on terrain, safety and environmental
concerns. The maximum loads are based on a minimum ice thickness of 107 centimetres (cm).
Planning for the next season needs to start early and involves projections of supply needs and
schedules for the upcoming year (Jarvis and Proskin 2011).
Nuna Logistics has been constructing and maintaining the TCWR since 1998. Along with
construction and maintenance, the company is responsible for truck dispatching, traffic control,
camp catering and summer camp maintenance.
The road was first built in 1982. Construction of the road generally begins in December or
January and takes approximately five to six weeks to complete. Historically, the road is open for
eight to ten weeks, starting anywhere from January 26 to February 11 and ending anywhere
from March 21 to April 16. In recent years it has opened before February 4 and closed by April
1. The average number of operational days per year is 67 (JVMC 2013).
The operating season for roads built over ice and compacted snow is sensitive to temperature
and other winter weather and climate conditions such as early- and late-season snowfalls
(Mesher, Proskin and Madsen 2008). Ice thickness and quality and overland portages are
particularly sensitive to rapid thawing (Comfort and Abdelnour 2013; Hayley and Proskin 2008;
McGregor, Hassan and Hayley 2008; Rawlings, Bianchi and Douglas 2009). The ongoing trend of
winter warming and shortened road seasons, as demonstrated in Appendix A, places resupply at
risk and also creates a perverse incentive to weigh operator safety against time and volume
constraints (McGregor, Hassan and Hayley 2008).
The majority of the road (approximately 87%) is built over lake ice and must be reconstructed
each year (JVMC 2013). The ice thickness is measured daily and profiled by ground-penetrating
radar. Ice sheet profiling is carried out throughout the season and the results are compared
with the data collected through quality assurance/quality control checks performed by an
engineering firm (JVMC 2013).
Construction of the road requires special lightweight equipment, including amphibious track
vehicles equipped with ground -penetrating radar. Helicopter surveillance identifies ice cracks
YUKON RESEARCH CENTRE
Case Study of the Tibbitt to Contwoyto Winter Road
6
or other hazards and supports construction with information as the machines move forward.
Snow cover insulates the ground or ice beneath it, so various types of low-ground-pressure
equipment — which are better able to float on soft ground surfaces — are used to keep the
road clear. This helps keep portages smooth and promotes the continual build-up of ice (Nuna
Logistics 2014).
The road is 50 metres (m) wide on lakes and 12 to 15 m wide on portages. Sections that cross
lakes are plowed wider than portages to ensure that the ice thickens in the driving lanes. Due to
the insulating effect of snow, ice underneath the snow banks can be thinner, weaker and may
be cracked (JVMC 2012). There are 65 overland portages between lakes, which are upgraded as
necessary with gravel and sand pads. In some years, a secondary winter road route has been
constructed by RTL Robinson Enterprises. It provides an alternate route for the southernmost
110 km.
Historical GPS coordinates are used to ensure that the TCWR crosses the deep areas of the lakes
while avoiding rocky shoals. Shoals are potential weaknesses, because deflection of the ice onto
shoals can erode the ice from below (JVMC 2013).
Road monitoring and maintenance are conducted throughout the season on both the primary
and secondary routes. This includes monitoring ice integrity, repairing cracks, filling potholes,
clearing snow, dealing with overflow and sanding portages.
2.4 ROAD TRAVEL
Trucks hauling fuel and supplies to the mines operate at a normal gross vehicle weight (GVW) of
63,500 kilograms (kg), with maximum loads of up to 100,000 kg (McGregor, Hassan and Hayley
2008). Load weight limits are based on the minimum ice thickness of the entire route. The
minimum ice thickness for hauling is 70 cm; only very light loads can travel at that point. As the
ice thickness increases, the allowable load weight rises commensurately. With an ice thickness
of 107 cm, a load of 42 t can travel on the ice. This is the equivalent of a Super B tanker fully
loaded with 50,000 litres (l) of fuel. In the past, the road allowed for a maximum of
approximately 800 loads per year, but with improved engineering the road can now handle up
to approximately 10,000 loads per year. Load numbers and ice thickness are recorded and
communicated to the public through the TCWR website (JVMC 2013).
EBA Engineering Consultants, Ltd. (now Tetra Tech EBA, Inc.) is one of the firms that provide
engineering support for the risk management program of the TCWR. These services include
measuring ice strength, analyzing ice-carrying capacity for standard and heavy loads, and
collecting data on ice thickness and bathymetry (Tetra Tech EBA 2013). The data collected by
FINAL REPORT
Economic Implications of Climate Change Adaptations for Mine Access Roads
in Northern Canada
7
Tetra Tech EBA is compared with that coming from Nuna Logistics to ensure that construction
and operation measurements are accurate.
NOR-EX Ice Engineering has provided engineering support for the TCWR since 2013, including a
project in 2014 that focused on ice engineering and quality assurance. The project reviewed ice
engineering procedures, developed new loading charts and conducted ice tests. This resulted in
fewer truckloads and increased payload capacity for the 2014 season (NOR-EX Ice Engineering
2014).
There are three maintenance camps along the road: Dome Lake, Lockhart Lake and Lac de Gras
(see Figure 1). The camps allow for the storing of materials, equipment and fuel, and provide
housing for maintenance and security staff. There is also a security check point at the Meadows
Dispatch at the beginning of the road. Drivers are required to comply with JVMC rules and
regulations, including speed limits, and a 24-hour patrol enforces those rules. Security is
provided by Deton’Cho/Scarlet Security Services, who deploy 15 to 18 officers based out of the
maintenance camps and provided with radar-based speed detection devices. Regular
inspections are also conducted by the Royal Canadian Mounted Police, federal and territorial
government representatives, and the JVMC (JVMC 2013).
The JVMC provides drivers with training and a written rule book, updated yearly, entitled
Winter Road Regulations and Rules of the Road. It provides information on speed limits and
other restrictions, and instructions for behaviour on the TCWR (JVMC 2012). The speed limit on
lakes is 25 km/hour when fully loaded, 35 km/hour when empty, and 10 km/hour for all weights
in a flood zone (damaged ice surfaces are repaired by being flooded). The speed limit on
portages is 30 km/hour, with a 10-km/hour speed limit for traveling on and off portages. Empty
trucks can drive 60 km/hour in “express lanes,” which are southbound return lanes built on
larger lakes. The secondary route is also sometimes used for southbound traffic. Speed limits
and spacing restrictions of 0.5 km between trucks are not only important for road safety, but
are also key elements in maintaining the integrity of the winter road ice surface and prolonging
the operating season (JVMC 2013).
3.0 ESTIMATING ECONOMIC IMPLICATIONS
The methodology used to evaluate the economic effects on the TCWR driven by climate change
employs the framework detailed in The Economic Implications of Climate Change on
Transportation Assets: An analysis framework (Sawyer 2014).This study combines the Sawyer
methodology with a conceptual framework developed specifically for the TCWR and described
in the steps below. Using this combination of established practice with a custom framework
ensures that analysis will be both rigourous and rooted in best practice.
YUKON RESEARCH CENTRE
Case Study of the Tibbitt to Contwoyto Winter Road
8
The following steps were undertaken to produce the analysis.
• Step 1: Hazard, Vulnerability and Assets at Risk. A literature review and Public
Infrastructure Engineering Vulnerability Committee (PIEVC) Engineering Protocol
(Engineers Canada 2014) were used to identify key relationships between climate, road
operations and adaptive responses.
• Step 2: Climate Conditions and Forecast. A PIEVC workshop was held in Yellowknife
with TCWR stakeholders to identify high-priority climate variables that affect road
operations. Forecasts for these weather variables were then developed using detailed
analytics.
• Step 3: Assessment Scenarios. Assessment scenarios were identified for estimating
economic outcomes of climate risks. Two scenarios were identified as being useful to
understand the climate-related links between weather and winter road operations: 1)
an Adaptive Scenario reflects operational changes that enable road operations to
continue, albeit at higher costs; and 2) a Critical Conditions Scenario reflects an
increased probability of more frequent road closures due to climate.
• Step 4: Economic Assets at Risk and Forecasts. The literature review, PIEVC process
outcomes, and interviews with road operators, users and government formed the basis
of the identification of key costs and their drivers.
• Step 5: Climate Risk and Economic Value Assessment. This step combined the climate
risks and the economic costs into relationships driven by weather variables. The
economic costs under the two scenarios were estimated using Monte Carlo analysis,
which uses probability distributions for all key variables to estimate the cumulative
uncertainty in each of the scenarios.
• Step 6: Net Costs of Climate-Induced Weather Pattern. An overall assessment of the
scenarios was made, highlighting the present value of the economic value at risk due to
climate change.
3.1 STEP 1: HAZARD, VULNERABILITY AND ASSETS-AT-RISK
3.1.1 Screening Assessment
The TCWR is located in an area of continental polar climate characterized by long cold winters.
Daily winter temperatures often fall below –30°C, and during the short cool summers
temperatures can reach 25°C (EBA 2001). Precipitation is sparse, with approximately half falling
as snow. Prevailing winds are from the east and average 15–17 km/hour. They often cause
FINAL REPORT
Economic Implications of Climate Change Adaptations for Mine Access Roads
in Northern Canada
9
blowing snow, which can be a problem for road operations (EBA 2001). Major storms with high
winds and blowing snow can cause temporary closures on the road, such as the storm
documented in March 2012 that caused a multiple-day road closure and subsequent road clean-
up (Rodan 2012). Storms such as the March 2012 event can endanger drivers because they
provide an incentive to reach a camp before the storm worsens, although some drivers end up
waiting out storms on portages (Rodan 2012). These unforeseen temporary road closures can
delay the road closing date if loads need to be hauled to the camps before the end of the
season (Rodan 2012). This implies that season closure dates for the TCWR may be influenced by
operational needs.
Trucks moving on the ice create waves of water under the ice surface, which causes it to flex.
Trucks driving over the wave can result in the ice breaking open, an event called a blowout
(Ashbury 2006). Blowouts can cause the road to be shut down for maintenance and, depending
on severity and timing, bring the risk of permanent closure. Limiting speeds and mandating
minimum spacing between trucks reduce the number of these ice blowouts. The roads are built
with S-curves at the portages to help limit speeds where the ice meets the shore. The road is
particularly vulnerable at these portage connections. Pressure ridges and cracks can be bridged
by steel ramps that vehicles travel on, which can avert road closure.
The JVMC has identified climate change and winter warming trends as a concern for the
longevity of the ice road (Greenspan 2008). Although 2007 was a record high year for loads,
2006 was one of the worst years and led the mine owners to start considering alternatives to
the ice road (Sherk 2007; McGregor, Hassan and Hayley 2008). The 2006 season opened late
and closed after only 49 days of operation because of unsafe ice conditions. The result was that
approximately 1,200 loads had to be flown into the mines in the summer and fall of 2006 at
great expense to the mining companies (JVMC 2013). Climate trends and their implications for
winter roads are further analyzed in Appendices A and B.
Following the short 2006 winter road season, a number of the short-term strategies mentioned
earlier were implemented that enhanced operations: (i) improved techniques for assessing ice
capacity and locating discontinuity using improved radar systems; (ii) traffic management
through express lanes to separate loaded trucks from returning trucks, allowing the speed
restrictions on the returning trucks to be relaxed; (iii) implementation of multiple routes across
lakes with known ice instability to allow rapid traffic redirection in the event of ice
deterioration; (iv) greater vigilance over speed restrictions, together with driver awareness
campaigns; (v) flooding ice roads and ice bridges to increase ice thickness and delay the closing
date; (vi) plowing snow off the road alignment; and (vii) restricting hauling to darkness hours
during warmer periods (McGregor, Hassan and Hayley 2008; Rawlings, Bianchi and Douglas
YUKON RESEARCH CENTRE
Case Study of the Tibbitt to Contwoyto Winter Road
10
2009). Building multiple routes has been identified as a key strategy for the TCWR and flooding
has been used with increasing frequency.
Similar short-term strategies have been employed on other winter roads. Spraying ice roads is
an alternative to flooding that is used frequently. Building a permanent bridge over road choke
points such as lakes and streams is another technique that has successfully extended the season
of other winter roads (Rawlings, Bianchi and Douglas 2009).
The climate analysis in Appendix A compared available climate data against operating season
length and found that the strongest indicator for longer seasons was the accumulation of
freezing-degree days, a winter temperature variable. Freezing-degree days have decreased
significantly over the past decades, indicating a tendency towards reduced ice thickness and a
shorter winter road season. Changes to road construction methods have mitigated that
outcome. The decrease in freezing-degree days has resulted in increased efforts to maintain ice
thickness and prolong the season. Projections indicate that this decrease in freezing-degree
days will continue. Other climate variables, such as amount of snowfall and rapid temperature
increases or decreases, were also investigated, but found to have less significance. Snowstorms
are less of a concern for the TCWR because there is a lot of equipment to maintain the road and
clear snow, but other winter roads may be more vulnerable to early-season snowfalls. If the
snow is not cleared in a timely manner, it may have enough of an insulating effect on the ice to
end the season.
Despite a prediction that the road would be viable for several decades (Natural Resources
Canada 2013), based on historical trends and projections of future climate change, changes in
ice road sustainability are likely. Freezing-degree days and melting-degree days will affect the
length of the season (Appendix B). Other research in the region is looking at historical
temperature trends, lake sediment cores and climate projections to get a better sense of how
climate change might influence the viability of the road (Galloway et al. 2010; Natural Resources
Canada 2013).
3.1.2 PIEVC Hazard Assessment
Hazard analysis identifies a specific set of circumstances that could potentially result in a
negative outcome. A hazard is a specifically defined interaction between a climatic event and a
component, or components, of a piece of infrastructure being studied. Using the PIEVC
protocol, climatic events were identified that could occur in the region within the time horizon
(to the 2050s) of the vulnerability assessment. These events were then applied against the
infrastructure components forming the TCWR to test how they would react; this identified a set
FINAL REPORT
Economic Implications of Climate Change Adaptations for Mine Access Roads
in Northern Canada
11
of hazards. The hazards were then ranked based on variables such as likelihood and severity,
which allowed a calculation of values at risk due to climate-derived events.
The PIEVC assessment considered 23 climate parameter/threshold combinations and 32
infrastructure components, yielding a total of 736 possible interactions. In consultation with
experts and technical advisors through interviews and a PIEVC workshop, 96 major
climate/infrastructure interactions were identified. These interactions form the core of the risk
assessment and are categorized according to severity:
• 12 high-risk interactions;
• 53 medium-risk interactions; and
• 31 low-risk interactions.
Risks were generally associated with five aspects of the TCWR:
1. Road operations. Road operations are vulnerable to climate conditions that could result
in load and service interruptions. While these interruptions can be made up through
road management practices in the early or mid-season, there may not be sufficient
operational management flexibility to recover if they occur later in the season.
2. Rapid temperature change: ice surface. Although rapid temperature change (greater
than 18°C in a 24-hour period) was considered to be a potential risk, subsequent climate
analysis indicated that rapid temperature changes are already being managed on the
road and that the probability of such events will slightly decrease over the time horizon
of the assessment. Therefore, rapid temperature changes were not considered to be a
significant climate change risk.
3. Pre-season snowfall: portages. Lack of pre-season snowfall can affect the preparation
of portages and ramps along the route. If the snow used to establish the portage
roadbed is in short supply, this will prolong preparation times and delay the opening of
the road.
4. Pre-season snowfall: thin ice. An abundance of pre-season snowfall was determined to
be a significant risk driver. Too much early-season snowfall can affect the rate of
freezing of the ice sheet. In addition, snow that covers thin ice represents a significant
safety hazard to crews preparing the road. Layers of snow can obstruct a clear view of
thin ice, resulting in machinery and personnel breaking through the ice surface.
5. Days above freezing: multiple infrastructure components. More frequent periods of
prolonged temperatures above freezing during winter operations could result in load
and service interruptions on the road.
YUKON RESEARCH CENTRE
Case Study of the Tibbitt to Contwoyto Winter Road
12
3.1.2.1 Road operations
In most areas, risks related to the normal road operation elements of the infrastructure
assessment. Since the operators accommodate for severe weather by adjusting operational
times, events that affect operational times tend to be a high priority. Events that reduce the
number of operational hours in the year can be important. Fortunately, the road is operated
based on total tonnage in a season and to date there has been sufficient flexibility in scheduling
in most years to accommodate changes in the operational season.
The winter of 2006 was an exception. The road experienced a blowout midway on Waite Lake,
which has an extremely rough bottom and many small islands and shoals (D. Hayley, pers.
comm.). The location that closed the road was between an island and the shore, where the road
passed over a shoal. In other locations the water under the ice was deep, but in this spot it was
only 30–40 cm deep. Despite slow speed limits, a hydrodynamic wave caused the blowout (D.
Hayley, pers. comm.). This happened fairly late in the season (March 14), which meant that an
alternate route could not be established. A significant amount of the seasonal hauling had not
yet been done and approximately 1,200 loads could not be delivered by the end of the season.
Participants in the hazard assessment workshop reported that the early closure of the TCWR
that year resulted in incremental costs between $100 and $150 million and contributed to the
temporary closure of the Jericho Diamond Mine in 2008 (it was sold in 2010 and reopened
briefly in 2012).
Of the 23 climate/threshold parameters initially identified by the assessment team, 7 were
ultimately removed from further analysis as they were deemed to be insignificant risk drivers in
relationship to road operations. Of the remaining 16 climate/threshold parameters, 11 yielded
medium- or high-risk scores for road operations. Based on this analysis, road operations were
found to have relevant risk interactions for roughly 70% of the climate parameters considered.
Generally there is sufficient flexibility in road operation scheduling to accommodate service
interruptions. Interruptions that are sufficiently severe, particularly late in the operating season,
can result in significant economic loss, as demonstrated during the 2006 season. For this reason,
road operations dominate the risk profile for the TCWR. Increasing the annual tonnage targets
for the road, while at the same time attempting to reschedule road operations around more
frequent climate-driven service interruptions, could exacerbate these risks.
3.1.2.2 Rapid temperature change/ice surface
The screening assessment identified rapid temperature change as a risk to road ice surface.
Rapid changes can trigger cracking in the surface ice sheet, which compromises the road’s
FINAL REPORT
Economic Implications of Climate Change Adaptations for Mine Access Roads
in Northern Canada
13
structural integrity. Consecutive days where these events occur increase the risk to the road ice
surface.
A climate parameter was developed that considered up to 20 consecutive days with rapid
temperature changes. It was established that these events already occur in the baseline climate
forecast and that they will likely decrease in frequency, if only slightly, over the time horizon of
the assessment.
These observations led to the conclusion that rapid temperature changes are currently being
managed by the JVMC and that the overall risk associated with rapid temperature changes over
the time horizon of the assessment will decrease based on the current and future climate
profile.
3.1.2.3 Pre-season snowfall/portages
Workshop participants identified high risks associated with the impact of pre-season snowfall
on portages. The road construction teams use early season snow to establish roadbeds through
the portages and to build ramps. A lack of early season snowfall can prolong the construction
period, as teams wait for sufficient snow to accumulate or are forced to move snow from lakes
onto the portages. This can delay the opening of portages, resulting in a shorter overall
operating season.
Given their associated risks, portages can be a vulnerable element of the ice road. The amount
of pre-season snowfall needs to be within a normal range to prevent unfavourable outcomes.
Deviations from this range can result in operational delays; as discussed in the section on road
operations, these are significant to the overall TCWR risk.
3.1.2.4 Pre-season snowfall/thin ice
Too much pre-season snowfall can also contribute to increased risk. The screening assessment
identified that too much snow in the early season slows ice thickening. Too much snow can hide
structural issues with the ice (both structural and surficial) and obscure the ability of workers
and equipment to detect hazards or thin ice, thereby contributing to safety risks. Historical
observations indicate that too much pre-season snowfall contributed to a hazardous event
during the construction phase of the 2003 road. Amphibious vehicles equipped with ground-
penetrating radar are now used to reduce risk while determining ice thickness.
Climate analysis indicates that periods of significant pre-season snowfall are likely to occur
during the horizon examined by this report. Since these events are highly likely, and since the
consequence in the worst case could be loss of life, they are deemed to be high risk. Although
YUKON RESEARCH CENTRE
Case Study of the Tibbitt to Contwoyto Winter Road
14
pre-season snowfall has implications for delays in road operations, it is first and foremost
evaluated for its potential effects on worker safety.
3.1.2.5 Days above freezing/multiple infrastructure components
A pattern of risk associated with days above freezing during the operating season was identified
through the analysis. Two threshold values were identified as being relevant:
• 0°C for three consecutive days; and
• 0°C for five consecutive days.
While no high-risk interactions were observed for these thresholds, 44 medium-risk interactions
were identified. As might be expected, risk scores were slightly higher for the five-consecutive-
day threshold, but both threshold values yielded a pattern of risk.
Prolonged periods of temperature above freezing could result in reductions in loads and service.
Operations might be possible only at night. Extended periods of above-freezing temperature
could shut down sections of the road. Late in the season, with longer days, temperatures above
0°C for even fewer than three consecutive days can cause quickly deteriorating conditions,
resulting in a temporary closure or a switch to night-time only use (R. Zschuppe, pers. comm.).
Consecutive days with above freezing temperatures are already occurring and, with climate
change, the team is projecting that it will be seen more often in the future.
Interactions have been observed in relation to multiple infrastructure components, with several
medium-risk components:
• ice surfaces;
• ice bridges;
• portages;
• vehicles;
• secondary spur roads; and
• road operations.
From this analysis, a general pattern of risk to the road is anticipated based on prolonged
periods of temperature above freezing. While much of this risk can be absorbed through flexible
scheduling, the days-above-freezing variable has the potential to drive high levels of risk if there
are enough of such days to significantly reduce the operating season.
FINAL REPORT
Economic Implications of Climate Change Adaptations for Mine Access Roads
in Northern Canada
15
3.1.3 Adaptation Responses
During the PIEVC assessment interviews and the workshop, experts discussed a number of
possible adaptation measures. Throughout the workshop, participants emphasized the
importance of flexible scheduling as a means of responding to climate fluctuations. For the most
part, this practice can result in minimal cost increases while achieving operational goals despite
service interruptions.
As indicated above, the JVMC has already invested in technologies to optimize the operability of
the winter road system, which in some years operates near or at capacity (McGregor, Hassan
and Hayley 2008). Initial economic studies by Babson College (2011) indicate that these
construction and maintenance techniques and technologies can lengthen winter road seasons,
saving between $6 million and $27 million a year in transportation costs. In the future, however,
such techniques may not be as effective. Climate change may reduce the operational season for
the road, resulting in less ability to reschedule. Additionally, the potential for new mines in the
region or increased annual tonnage requirements could further compromise scheduling
flexibility. Indeed, the initial modelling by Babson College indicated that in the future the TCWR
may fail to deliver sufficient supplies in some winters and that the additional cost for each failed
year increases over time. The Babson study concluded that given the effects of climate change
future operations may require alternative transportation methods (Babson College 2011).
Nevertheless, in the short term, flexible scheduling is a bridge between the current mode of
operation and other more robust adaptation measures. As part of this bridging strategy, it will
be important to focus on key maintenance and operational practices, including monitoring of
the road components and weather, and ongoing analysis of climate projections for the region.
The road operators are already applying these processes, so there should be minimal financial
costs associated with continuing and enhancing these activities.
Other, more robust adaptation measures were discussed by workshop participants:
• Construction of permanent bridges. One expert proposed the use of permanent bridge
structures as an adaptation for ice roads; these would be used only during the winter
road operational period. Workshop participants indicated that there are no river
crossings on the current TCWR alignment. Thus, this adaptation measure may not be
relevant in the present analysis. However, it may be viable for other ice road
applications.
• Construction of permanent road alignments on portages. Some experts suggested that
building permanent road alignments on portages would be a potential adaptation
measure. In this scenario, permanent roadbeds would be prepared, but used only when
YUKON RESEARCH CENTRE
Case Study of the Tibbitt to Contwoyto Winter Road
16
the ice road was in service. This would eliminate the sensitivity to early season snowfall
and allow road operations to commence as soon as the ice sheet was deemed safe.
Permanent road padding, made up mostly of esker sand, has already been put in place
in some sections.
• Construction of permanent all-season roads. If winter road operations become
seriously affected by changing climate conditions, some of the experts suggested a
permanent all-season road. This option could be built in phases, constructing
permanent all-season roads where feasible and connecting with ice roads during winter
operations where necessary.
The construction of an overland route — along with construction of a deep sea port and
bringing in power lines to reduce fuel needs — were some of the large-scale adaptation
responses that the JVMC considered in 2007, among many other potential options. The cost,
time frame and environmental issues of each option were analyzed to identify the most realistic
alternatives (Finlayson 2007).
Three final concepts were considered, all of which would take approximately five years to
implement and all of which were expected to be very costly. In 2007, when these concepts were
being considered, annual loads were projected to increase up to 14,000 t, which would likely be
more than the existing road could handle (Finlayson 2007; Sherk 2007).
• Construction of a seasonal overland route. This would parallel the TCWR along its most
southern portion. This is the section that melts earliest and was the source of most of
the 2006 difficulties. The new road would be a gravel-surface route, built on a base of
shot rock fill — a foundation of broken stones often used in water or on soft ground —
with a smooth gravel layer on top, and covered by compacted snow. Travel on the road
would not be permitted until the snow base had been built up to protect the fill
material.
• Construction of a deep sea port. Another proposed option was to build a road from a
deep sea port at Bathurst Inlet; the port has been proposed, but not yet built. This
option would benefit Nunavut, provide a route for base metals as well as diamonds, and
be another way for the federal government to exert Canadian sovereignty in the Arctic.
• Bringing in power lines. The third alternative was to build 600 km of power lines from
the Taltson hydroelectric station near Fort Smith and expand the station’s production.
This would reduce the reliance of the mines on diesel power, and since diesel is the
main good transported to the mines, this would greatly minimize loads.
FINAL REPORT
Economic Implications of Climate Change Adaptations for Mine Access Roads
in Northern Canada
17
The construction of a seasonal overland road as an adaptation option along the southern
portion of the road was reiterated in the expert interviews following the PIEVC workshop. The
Government of the Northwest Territories (GNWT) has recently revived its interest in
constructing a southern overland road to the Lac Des Gras area of the NWT (Quenneville
2015a). The GNWT is considering financing options for an all-weather road to improve
transportation access into the Slave Geological Province (SGP) to better support diamond
mining operations and to encourage new mineral exploration and mine development in this
mineral-rich region.
Investment in extending the NWT’s all-weather road system into the SGP would significantly
reduce transportation-related costs to operating mines and extend mine life. It would limit the
effects of climate change to spur ice roads to mining installations (if construction of such spur
roads is continued under an all-weather access model). It would also significantly reduce the
annual cost of constructing and maintaining winter roads that serve mining-related projects in
the NWT and western Nunavut. An all-weather road would not eliminate all climate change risk,
since climate still influences roadways, but it would reduce significantly affect the operating and
cost implications of relying on seasonal, ice road access (P. de Bastiani, Assistant Director of
Planning, GNWT Department of Transportation, pers. comm.).
3.2 STEP 2: CLIMATE CONDITIONS AND FORECAST
This step in the methodology aims to quantify the economic implications of projected climate
trends on the TCWR’s operation and use. This will inform the rationale for some of the more
robust adaptation measures that are being considered.
These climate variables were identified as important through the PIEVC workshop and further
climate analysis undertaken by Risk Science International in relation to the economics of the
TCWR:
• operational season length (interaction of freezing-degree days and melting-degree
days);
• incidence of temperature swings in excess of 18˚C;
• incidence of consecutive days above 0˚C during the operational season;
• amount of snow on the ground January 1; and
• number of extreme cold events during the operational season.
Projections for these variables’ future expected values and their associated probability
distributions are presented in Figure 2. Estimates for these climate variables are drawn from the
RCP8.5 scenario in the Intergovernmental Panel on Climate Change (IPCC) assessment reports
(IPCC 2013).
YUKON RESEARCH CENTRE
Case Study of the Tibbitt to Contwoyto Winter Road
18
Figure 2: Emission projections from IPCC modeling
Source: Fuss et al. 2014
Perhaps the most convincing argument for the use of RCP8.5 climate scenarios in the projection
of relevant climate variables for this project, versus RCP4.5 or RCP6 or the blending of scenarios,
is that based on historical trends, the planet is definitively following the RCP8.5 pathway, a
trend that has continued in the 2014 data. In addition, given the lack of binding international
agreements on greenhouse gas (GHG) emissions and the atmospheric persistence of GHGs
(hundreds of years), present trends are unlikely to be reversed between now and 2050.
Therefore the RCP8.5 level of climate change is a reasonable expectation for the study period.
There have been recent indications that a global GHG agreement may be possible by 2020. The
United States and China have recently announced a bilateral agreement that they hope can lead
to a global agreement. When examined closely, however, the proposed commitments are not at
a level that would alter projections for the 2050s. For this reason — and because of the need to
engineer TCWR road infrastructure to be risk-resilient — the projections below are consistent
with the RCP8.5 scenario.
FINAL REPORT
Economic Implications of Climate Change Adaptations for Mine Access Roads
in Northern Canada
19
Estimates are provided for climate variable values in the 2020s and 2050s. For the purpose of
calculations, a linear trend line between the two periods was applied, since such a trajectory is
consistent with the projected evolution of variable values.
In terms of these variables’ probability distributions, the preferred approach would be to apply
historic distributions to estimates of future viability and then add to this an additional standard
deviation factor associated with the range of future climate change projections. However, this
second step cannot be easily accomplished without applying a 24-hour time period to the
calculations in the climate models, which were carried out monthly or seasonally. Therefore, in
order to account for the added uncertainty in estimates of future climate conditions, the
relative magnitude of the range in variables’ values that has been historically observed has been
taken as the range for climate variables’ future distributions. In terms of the probability
distribution within this range, a normal distribution is used; it applies the same standard
deviation that was experienced historically, but as a proportional share of the given time
period’s expected value.
3.2.1 Projected Climate Variables
3.2.1.1 Operational season length
A possible correlation was evaluated between warm early seasons and warm road seasons
overall to better understand the impact of temperature on operational season length. This was
done by taking the average of all December to January (Dec–Jan) mean temperatures from
1943–2013 and the average of all December to March (Dec–Mar) mean temperatures for the
same years. Each year’s value was then compared to the mean to see if it was a positive
anomaly (warmer than average) or negative anomaly (colder than average). The results clearly
indicated that if the Dec–Jan period was colder than normal, the entire season was colder than
normal and if Dec–Jan was warmer than normal, the entire season was warmer than normal.
This can be seen in Figure 3, where positive anomalies and negative anomalies per year are
almost always synchronized. This implies that most of the time (61 of the 71 years calculated, or
86% of the years); the Dec–Jan temperature is a good indicator of the entire season (Dec–Mar)
temperature. The ten years where the Dec–Jan and Dec–Mar temperature anomalies did not
match were 1948, 1955, 1967, 1968, 1977, 1978, 1979, 1984, 2005 and 2008. In most of these
non-matching years, the anomalies were still close to agreement.
YUKON RESEARCH CENTRE
Case Study of the Tibbitt to Contwoyto Winter Road
20
Figure 3: Correlation between warm early season (Dec–Jan) and warm road season overall
(Dec–March) using mean temperature anomaly at Yellowknife A
Data source: Environment Canada 2014
Under historical climate conditions, the TCWR is generally resilient. However, the road may not
be as resilient over the assessment time horizon of the 2020s and 2050s (Table 1). Over these
periods, the average operating season duration may be reduced to 41 days. An exceptionally
warm year during a period with an average season length of 41 days could easily reduce the
operational season to around 30 days, based on the forecast climate analysis. This situation
could lead to increased difficulties in applying the flexible scheduling methodologies currently
used and could result in failure to meet annual tonnage targets. Table 1 compares the historic
and projected operational season lengths (JVMC 2013) based on the climate analyses in
Appendices A and B.
Table 1: Operational season length (days)
Time Period
Value/estimate
(Number of days)
Range
(Number of days) Standard Deviation
1981–2010 65 47–78 +/–8
2020s 58 44–72 +/–7
2050s 49 37–61 +/–6
-8
-6
-4
-2
0
2
4
6
8
10
1943
1945
1947
1949
1951
1953
1955
1957
1959
1961
1963
1965
1967
1969
1971
1973
1975
1977
1979
1981
1983
1985
1987
1989
1991
1993
1995
1997
1999
2001
2003
2005
2007
2009
2011
2013
Dec-Jan Anomaly Dec-Mar Anomaly
FINAL REPORT
Economic Implications of Climate Change Adaptations for Mine Access Roads
in Northern Canada
21
3.2.1.2 Incidence of temperature swings in excess of 18°C
Temperature swings are not expected to become more or less prevalent over the forecast
period. Their probability distribution is also expected to remain unchanged from the historical
average (Table 2).
Table 2: Incidence of temperature swings in excess of 18°C (events per season)
Time Period
Value/estimate
(Events/season) Range Standard Deviation
1981–2010 31 18–55 +/–10
2020s 31 18–55 +/–10
2050s 31 18–55 +/–10
3.2.1.3 Incidence of consecutive days of temperatures above 0°C during operational season
The likelihood of consecutive days of ice-melting temperatures during the operational season is
expected to increase significantly over the forecast period. Such events are sufficiently rare that
even with this increased likelihood the expected incidence is still fairly low, with only one three-
day consecutive event expected during the operational season by the 2050s. Probability
distributions were not available for these figures; therefore, a range and standard deviation of
zero was applied (Table 3).
Table 3: Number of consecutive days > 0°C, November 1–April 1
Time Period
Three-day consecutive days >°C Five-day consecutive >°C
Events/year Range
Standard
deviation Events/year Range
Standard
deviation
1981–2010 0.10 0 0 0.00 0 0
2020s 0.27 0 0 0.03 0 0
2050s 1.13 0 0 0.40 0 0
3.2.1.4 Snow on ground January 1
The amount of snow on the ground January 1 affects both ice formation and safety
considerations when constructing the winter road and the feasibility of portage construction. It
is projected to decline slightly over the study period (Table 4). The impacts of this variable will
be felt more in portage construction, since it requires snow.
YUKON RESEARCH CENTRE
Case Study of the Tibbitt to Contwoyto Winter Road
22
Table 4: Snow on ground January 1 (cm)
Time Period Value/estimate (cm) Range (cm) Standard Deviation
1981–2010 28.2 4–61 +/–10.5
2020s 23.8 0–51 +/–8.9
2050s 20.9 0–45 +/–7.8
3.2.1.5 Incidence of extreme cold events during operational season
The incidence of extreme cold events is expected to decrease significantly over the forecast
period (Table 5), meaning that costs associated with coping with extreme cold will likely fall
over time.
Table 5: Incidence of extreme cold events (<–35°C) during the operational season
Time Period Value/estimate Range Standard Deviation
1981–2010 24.27 4–50 +/–11.3
2020s 16.17 1–31 +/–7.5
2050s 6.37 0–12 +/–3
3.3 STEP 3: ASSESSMENT SCENARIOS
Based on the vulnerability assessment and the detailed climate data that was assembled, two
scenarios were identified that guided the analysis:
1. Operations can adapt to a changing climate (adaptive scenario). In this scenario,
increasingly difficult climate conditions lead to shorter operational seasons and steadily
increasing maintenance and repair requirements to maintain road service levels. Road
operations remain functional through adaptive measures.
2. Operations are disrupted due to a changing climate (critical conditions scenario). In
this scenario, highly challenging climate conditions lead to significantly increased costs
due to the late opening and/or early closure, or non-opening, of the TCWR.
These two scenarios capture the two basic outcomes possible under increasingly difficult
climate conditions on the TCWR: operation and maintenance become more difficult and costly;
or the desired levels of road availability become impossible, necessitating the use of alternative
transportation measures or making it impossible to move some loads altogether.
FINAL REPORT
Economic Implications of Climate Change Adaptations for Mine Access Roads
in Northern Canada
23
Thresholds for the climate variables outlined above were identified through consultation with
stakeholders. This information indicated which specific kind of climate conditions, and what
type of interactions between these conditions, would be associated with each of the two
scenarios.
3.4 STEP 4: ECONOMIC ASSETS-AT-RISK AND FORECASTS
This step links the detailed climate analytics with the likely costs that are expected under the
two assessment scenarios. The approach uses welfare economic principles2 to estimate the
incremental economic costs under each scenario, all of which are estimated using Monte Carlo
analysis to factor for variable uncertainties. A Monte Carlo analysis is commonly used to model
phenomena with significant uncertainty. Probability distributions were used for all key variables
to estimate the cumulative uncertainty in each of the scenarios. All key input parameters are
expressed as probability density functions. More technical detail on the approach and methods
can be found in Sawyer 2014.
The focus of this step is to identify the significant costs, estimate their magnitude and
associated uncertainty, and link them with the relevant climate variables. The objective is to
understand how changes in climate variables are expected to drive changes in cost variables.
(Step 5 extends this analysis in order to establish exactly how and why costs are affected.)
Implicit in Steps 4 and 5 is the development of a damage function, an equation that specifies
the relationships between costs and climate variables and thereby establishes how much
economic cost is associated with a given change in one or more climate variables.
Data collection on economic assets at risk was undertaken through a combination of a literature
review, interviews and surveys with road operators and users. Literature on ice roads was
reviewed to develop estimates of the various costs considered in this analysis and these costs
were verified or discussed by participants in the PIEVC protocol workshop and other TCWR
stakeholders. Phone interviews were conducted with a range of stakeholders, some of whom
received a follow-up questionnaire that asked them to provide background information, cost
figures and estimates of likely cost impacts under current and worsening climate scenarios.
Table 6 provides an overview of variables that were found likely to trigger economic costs under
various climate scenarios. These costs form the basis of developing the economic losses
associated with a changing climate.
2 Welfare economics uses microeconomic techniques to evaluate well-being at the economy-wide level. It provides
the basis for public economics and tools such as cost-benefit analysis.
YUKON RESEARCH CENTRE
Case Study of the Tibbitt to Contwoyto Winter Road
24
Table 6: Cost impacts triggered by a changing climate
Cost variables Description Impacted stakeholder(s)
Flexible scheduling costs
(shorter season)
Logistical costs and personnel
and capital costs associated with
coping with a late start and/or
early end of the operational
season
Carriers
Flexible scheduling costs
(interruptions to operations)
Logistical costs and personnel
and capital costs associated with
interruptions to road availability
during the operational season
Carriers
Ice thickening or repair measures
(flooding and sanding)
The costs of thickening ice, either
to build up an insufficient
amount or to build up ice on
sections that are in need of
repair
Road operators
Increased ice road construction
and maintenance costs
The increase in ice road
construction and maintenance
costs due to changes in climate
conditions during the
operational season
Road operators
Increased portage construction
and maintenance costs
The increase in portage
construction and maintenance
costs due to changes in climate
conditions during the
operational season
Road operators
Increased ramp construction and
maintenance costs
The increase in ramp
construction and maintenance
costs due to changes in climate
conditions during the
operational season
Road operators
Safety-related costs associated
with working on thin ice
The cost, over and above the
baseline, of safety-related
responses to the need to work
on thin ice
Road operators
FINAL REPORT
Economic Implications of Climate Change Adaptations for Mine Access Roads
in Northern Canada
25
Cost variables Description Impacted stakeholder(s)
Safety and equipment failure
costs dues to extreme cold
The cost, over and above the
baseline, of safety-related
responses and equipment
failures associated with an
increasing incidence of -40°C
temperatures
Road operators
Modal shifting The cost associated with the use
of alternative transport
measures when the winter road
season is not sufficient to meet
demand
Mines
Production loss Lost economic value associated
with production that did not
occur due to an inability to
transport needed supplies and
equipment during a season
Mines
As mentioned above, estimates of the magnitude of the climate-driven changes in these cost
variables were developed through surveys, interviews and a literature review. Table 7 provides
an overview of the cost parameters used in the analysis, which are summarized using minimum,
central and maximum values. In some cases a minimum and maximum value were identified
directly; in other cases, an assessment of the range of uncertainty associated with the central
figure was used to develop the minimum and maximum values. When uncertainty information
was unavailable, qualitative uncertainty estates were assigned quantitative uncertainty ranges:
a variance of either +/– 5%, 15%, or 40% around the central value was applied to estimates with
low, medium, and high uncertainty, respectively. A triangular distribution3 among minimum,
central and maximum values was employed in the ultimate analysis.
Table 7: Cost parameter values used in economic analysis
Cost Variable Units Minimum Central Maximum Source
Flexible scheduling costs
(shorter season) $/tonne-km $0.07 $0.12 $0.18
RTL,
Prentice,
2013
3 A triangular distribution is a continuous probability distribution with a lower limit a, upper limit b and mode c; a <
b and a ≤ c ≤ b. It is used when the relationship between variables is known but data is scarce.
YUKON RESEARCH CENTRE
Case Study of the Tibbitt to Contwoyto Winter Road
26
Cost Variable Units Minimum Central Maximum Source
Flexible scheduling costs
(interruptions to operations) $/tonne-km $0.03 $0.06 $0.09
RTL,
Prentice,
2013
Ice thickening or repair
measures (flooding and sanding) $/hour $550 $650 $750 Nuna
Logistics
Increased ice road construction
and maintenance costs $/hour $170 $200 $230 Nuna
Logistics
Increased portage construction
and maintenance costs $/hour $106.25 $125 $143.75 Nuna
Logistics
Increased ramp construction and
maintenance costs $/hour $63.75 $75 $86.25 Nuna
Logistics
Safety related costs associated
with working on thin ice $/hour $170 $200 $230 Nuna
Logistics
Safety and equipment failure
related costs due to extreme
cold
$/hour $63.75 $75 $86.25 Nuna
Logistics
Modal shifting costs $/tonne $576 $960 $1344 Various
sources
Production loss costs $/tonne $1350 $2250 $3150 Various
sources
With the exception of costs related to modal shifting (using alternative transport measures
when required loads cannot be transported via winter road) and production loss, all central
estimates for the cost variables in Table 7 were collected via interviews and surveys. Flexible
scheduling costs were estimated using a baseline estimate of northern ice road transport costs,
which averaged $0.32 per tonne-kilometer (Prentice 2013). These and other cost-benefit
analysis inputs and assumptions are summarized in Table 8. Where distributional information
was not available for the figures seen in Table 8, the same 5%, 15%, and 40% variances were
applied, based on the assessed level of associated uncertainty, and triangular distributions were
again applied in the ultimate analysis.
FINAL REPORT
Economic Implications of Climate Change Adaptations for Mine Access Roads
in Northern Canada
27
Table 8: Inputs and assumptions used in economic analysis
Cost Variable Units Minimum Central Maximum Source
Ice road transport cost $/tonne-km 0.272 0.32 0.368 Prentice
2013
Alternative transport cost $/tonne-km 1.92 3.2 4.48 Quenneville
2015b
Average distance traveled km 255 300 345 Estimated
Average load tonnage tonnes 27.58 32.45 37.31 JVMC 2014
Average seasonal tonnage tonnes/seas
on 120,020 21,6643 33,0002 JVMC 2014
Road length km 380 400 420 JVMC 2014
Ice road length km 323 340 357 JVMC 2014
Portage length km 57 60 63 JVMC 2014
Number of portages (no units) 61.75 65 68.25 JVMC 2014
Number of ramps (no units) 123.5 130 136.5 JVMC 2014
Ice road repair time (1/4 of road
length) hours 340 567 793 Estimated
Total ice road construction time hours 867 1,020 1173 Estimated
Total portage construction time hours 1,530 1,800 2070 Estimated
Total ramp construction time hours 663 780 897 Estimated
Costs due to modal shifting and production loss were estimated by examining the 2006 failure
of the TCWR.4 In that year, approximately 1,200 loads did not travel via the ice road (JVMC
2014). By examining TCWR road statistics from 2002–2012 it was determined that the average
load during this time weighed 32.454 t (JVMC 2014); therefore, the weight of the 1,200 loads
could be estimated at 38,945 t. The 2006 closure was estimated by road operators to have cost
road users $100–150 million, so the central estimate of $125 million was divided by the
estimated total tonnage of 38,945 for the 1,200 loads to produce an expected combined modal
shifting and production loss cost of $3,210 per tonne (assuming an average distance traveled of
4 This method was selected because no estimates or projections of future closure costs emerged from the project’s
interviews, survey or literature review. The historical experience was therefore determined to be the most reliable
means of estimating potential future costs. Because of the uncertainty surrounding this estimate the maximum
variance of 40% was applied.
YUKON RESEARCH CENTRE
Case Study of the Tibbitt to Contwoyto Winter Road
28
300 km). This figure was then broken out into its component parts (both modal shifting costs
and production losses) by assuming that alternative transport measures are typically ten times
as costly (Quenneville 2015b) and by again assuming that the average distance traveled was 300
km. Based on these calculations, the cost of transporting the 38,945 t in 2006 by alternative
means was estimated to be approximately $960 per tonne. The balance of the combined figure
for modal shifting and production cost was then attributed to the production loss component,
giving an estimate of approximately $2,250 per tonne for this figure. This method thereby
estimates a 3:7 ratio between modal shifting and production costs.
With respect to interactions between cost and climate variables, Table 9 provides an overview
of the climate variables identified as being relevant to road operation and use costs, the
conditions under which they have an impact, the type of costs they incur and the stakeholders
they affect under these conditions. The basis for many of these relationships and anticipated
impacts emerged from the PIEVC workshop and from subsequent expert interviews. The effects
of climate variables on road operation and use (and thereby costs) is discussed in detail in
Section 3.2.
Table 9: Changing climate variables and affected costs
Changing climate
variables
Impact scenarios and their
defined thresholds
Affected cost
types
Affected stakeholder(s)
Carriers Road
operators Mines
Operational season
length
Operational season
shortens due to
interaction of the
number of freezing-
degree days and
melting-degree days
Adaptive Scenario
Threshold: Operational
season <60 days
Flexible
scheduling costs
(shorter season)
— —
Critical Conditions Scenario
Threshold: Operational
season <50 days
Modal shifting — —
Production loss — —
Incidence of
temperature swings
in excess of 18°C
Can lead to ice
cracking and shifting,
which compromises
road’s integrity
Adaptive Scenario
Threshold: three or more
incidences
Flexible
scheduling costs
(interruptions to
operations)
— —
Ice thickening or
repair measures
(flooding and
sanding)
— —
FINAL REPORT
Economic Implications of Climate Change Adaptations for Mine Access Roads
in Northern Canada
29
Changing climate
variables
Impact scenarios and their
defined thresholds
Affected cost
types
Affected stakeholder(s)
Carriers Road
operators Mines
Critical Conditions Scenario
Threshold: Interaction with
other conditions
Modal shifting — —
Production loss — —
Incidence of
consecutive days of
above 0°C
temperatures during
operational season
Prolonged periods of
temperature above
freezing during
operational season,
necessitating repair
to sections of the
road
Adaptive Scenario
Threshold: two occurrences
of two consecutive days or
one occurrence of five
consecutive days
Flexible
scheduling costs
(interruptions to
operations)
— —
Ice thickening or
repair measures
(flooding and
sanding)
— —
Critical Conditions Scenario
Threshold: Interaction with
other conditions
Modal shifting — —
Production loss — —
Snow on ground
January 1 – ice road
impacts
Too much snow on
the ice in the early
season slows ice
thickening and can
create safety
concerns during ice
road construction
Adaptive Scenario
Threshold: More than 25
cm
Increased ice
road
construction and
maintenance
costs
— —
Critical Conditions Scenario
Threshold: Interaction with
other conditions
Modal shifting — —
Production loss
— —
Snow on ground
January 1 – portage
and ramp impacts
Too little pre-season
snowfall can delay
the opening of
portages since ramps
need snow for
construction and
Adaptive Scenario
Threshold: Less than 15 cm
Increased ramp
construction and
maintenance
costs
— —
Increased
portage
construction and
maintenance
costs
— —
YUKON RESEARCH CENTRE
Case Study of the Tibbitt to Contwoyto Winter Road
30
Changing climate
variables
Impact scenarios and their
defined thresholds
Affected cost
types
Affected stakeholder(s)
Carriers Road
operators Mines
portages require
sufficient snow to
meet licensing
requirements
Critical Conditions Scenario
Threshold: Interaction with
other conditions
Modal shifting — —
Production loss — —
Incidence of
extreme cold events
during operational
season
Extreme cold of
temperature below –
40°C can have
implications for
worker safety and
equipment function
Adaptive Scenario
Threshold: More than eight
occurrences
General increase
in operational
costs
—
Critical Conditions Scenario
(N/A)
As discussed in Section 3.3, the adaptive scenario describes a situation in which increasingly
difficult conditions are encountered, but where road operators and users can adjust under
slightly higher costs. The critical conditions scenario describes a situation where the road is not
able to meet the demand, despite adaptation measures. As seen in Table 9, a number of climate
conditions are considered in the adaptive scenario. The critical conditions scenario is triggered
only under circumstances that lead to an operational season that is historically unprecedented
in its shortness. Table 9 shows how the scenarios are triggered by changes in various climate
variables, and which costs are affected when climate scenarios are triggered. The specific
magnitudes of these costs under various climate scenarios and conditions are described in
Section 3.5.
3.5 STEP 5: CLIMATE RISK AND ECONOMIC VALUE ASSESSMENT
This step combines the outputs of the previous steps to produce an assessment of climate and
economic value at risk. This is done by linking climate conditions and scenarios with specific cost
impacts. Step 5 merges the climate risks with the knowledge of present costs and their
expected increases if particular climate conditions worsen.
Tables 10 and 11 convey the specific increases to costs under the two types of climate
scenarios: adaptive and critical conditions. Baseline cost data is also provided for context. For
example, for flexible scheduling costs, the cost increment associated with a shorter operational
FINAL REPORT
Economic Implications of Climate Change Adaptations for Mine Access Roads
in Northern Canada
31
season is $0.12 per tonne-km, which is over and above the baseline cost of $0.32 per tonne-km.
The table shows the central values for each cost variable as summarized in Table 7; the analysis
uses the probability distributions seen in Tables 7 and 8.
As seen in Table 10, an operational season of less than 50 days is taken as the threshold for an
adaptive scenario. Respective snowfall-related thresholds for this scenario of less than 15 and
greater than 25 cm were drawn from interviews. Other threshold values were for the most part
determined by looking at what would be above the historical average. In all cases these
thresholds are deviations from the norm that interview and survey respondents believed could
be dealt with, but that in some cases would involve higher costs.
Table 10: Incremental costs under adaptive scenario climate conditions
Cost type Climate threshold Impact Baseline Incremental
cost
Source(s)
Flexible
scheduling
costs (shorter
season)
Operational season
length <60 days
25-50% increase in
transport costs
$0.32 per
tonne-km
$0.12 per
tonne-km
RTL, JVMC,
Prentice,
2013
Flexible
scheduling
costs
(interruptions
to
operations)
35 or more incidences
of temperature swings
in excess of 18°C
Marginal impact if
temperatures
< 0°C
$0.32 per
tonne-km
Negligible
RTL, JVMC,
Prentice,
2013
Two or more
occurrences of three
consecutive days of
above zero
temperatures
10-25% increase in
transport costs due
to associated road
operation
suspensions
$0.06 per
tonne-km
RTL, JVMC,
Prentice,
2013
Ice thickening
or repair
measures
(flooding and
sanding)
35 or more incidences
of temperature swings
in excess of 18°C
Marginal impact if
temperatures
< 0°C $550-$750
per hour,
0.1-0.2 km
per hour
Negligible Nuna
Logistics
Two or more
occurrences of three
consecutive days of
above zero
temperatures
Mid-season flooding
operations triggered
$650 per
hour,
affecting ¼
of ice road
length
Nuna
Logistics
YUKON RESEARCH CENTRE
Case Study of the Tibbitt to Contwoyto Winter Road
32
Cost type Climate threshold Impact Baseline Incremental
cost
Source(s)
Five or more
incidences of extreme
cold events during
operational season
Costs at high end of
baseline range
$100 per
hour higher
than
average
cost
Nuna
Logistics
Ice road
construction
and
maintenance
Snow on ground
January 1 >25cm
Road construction
costs at high end of
baseline range due to
safety concerns
associated with
heavy snow on thin
ice formation
$750-
$1,000 per
hour,
0.33 km
per hour
$125 per
hour higher
than
average
cost
Nuna
Logistics
Five or more
incidences of extreme
cold events during
operational season
Costs at high end of
baseline range
$125 per
hour higher
than
average
cost
Nuna
Logistics
Portage
construction
and
maintenance
Snow on ground
January 1 <15cm
Portage construction
costs at high end of
baseline range due to
need to use lake
snow
$450-$600
per hour,
33 metres
per hour
$75 per
hour higher
than
average
cost
Nuna
Logistics
Five or more
incidences of extreme
cold events during
operational season
Costs at high end of
baseline range
$75 per
hour higher
than
average
cost
Nuna
Logistics
Ramp
construction
and
maintenance
Snow on ground
January 1 <15cm
Ramp construction
costs at high end of
baseline range due to
need to use more
lake snow
$1,400-
$1,800 per
hour,
six hours
per ramp
$200 per
hour higher
than
average
cost
Nuna
Logistics
FINAL REPORT
Economic Implications of Climate Change Adaptations for Mine Access Roads
in Northern Canada
33
Cost type Climate threshold Impact Baseline Incremental
cost
Source(s)
Five or more
incidences of extreme
cold events during
operational season
Costs at high end of
range
$200 per
hour higher
than
average
cost
Nuna
Logistics
An operational season length of less than 45 days was taken as the basis for triggering a critical
conditions scenario (Table 11), since at that point the road would not be able to accommodate
an average season’s demand, even if road operators worked at their historical peak efficiency. A
critical conditions scenario stemming from an interaction of climate variables was also specified;
specifically, an operational season length of less than 50 days combined with more than two
incidences of consecutive days of above zero temperatures, more than 25 cm of snow on the
ground January 1, or less than 15 cm of snow on the ground January 1. This alternative scenario
was specified to capture the risk inherent if a relatively short season combined with operational
complications to produce an even shorter season.
Table 11: Incremental costs under critical conditions scenario
Cost type Climate threshold Impact Baseline Incremental
Cost Source(s)
Modal
shifting
Operational season
length <45 days
Short operational
season window
necessitates use of
alternative transport
cost
N/A
(purely
increment
al cost)
$960 per
tonne
Various
sources
(see
Section
4.4)
Interaction of
conditions -
temperature swings,
above zero
temperatures and
irregular snow
conditions
Interruptions to
service due to
adverse climate
conditions
necessitates use of
alternative transport
measures, such as
airlifts
$960 per
tonne
Production
loss
Operational season
length <45 days
Short operational
season window
causes key pieces of
equipment/infrastru
N/A
(purely
increment
$2250 per
tonne
Various
sources
(see
Section
YUKON RESEARCH CENTRE
Case Study of the Tibbitt to Contwoyto Winter Road
34
Cost type Climate threshold Impact Baseline Incremental
Cost Source(s)
cture to not be
delivered
al cost)
4.4)
Interaction of
conditions -
temperature swings,
above zero
temperatures and
irregular snow
conditions
Interruptions to
service due to
adverse climate
conditions causes
key pieces of
equipment/
infrastructure to not
be delivered
$2250 per
tonne
3.6 STEP 6: NET COSTS OF CLIMATE-INDUCED WEATHER PATTERNS
In Step 6, the net costs were estimated for each of the two scenarios by combining the
estimates and the distribution of specific costs (see Table 7), specific climate variables (Section
4.2), and their interactions (discussed in Sections 4.4 and 4.5). The resulting net cost figures
were broken down using the cost typology identified in Table 6. The results of this analysis are
presented in Section 4.
4.0 RESULTS
The results of both scenarios are provided in Table 12. For the adaptive scenario, where the
mines and the road operator conduct ongoing actions to adjust to varying climate change, the
greatest cost results from adaptive scheduling on the part of carriers. To the extent that the
season is shortened below the threshold of 50 days, additional costs are triggered as loads are
shifted. In total, over the assumed 35-year time horizon, the total costs of the adaptive scenario
are in the order of $55 million.
Table 12: Scenario results ($ millions NPV @ 4%: 35 years)
Adaptive scenario 20th
percentile Mean 80th
percentile
Flexible scheduling costs (shorter season) $28.45 $44.26 $58.64
Increased ice road construction and maintenance
costs $5.18 $5.77 $6.35
Increased portage construction and maintenance
costs $4.69 $5.26 $5.79
FINAL REPORT
Economic Implications of Climate Change Adaptations for Mine Access Roads
in Northern Canada
35
Adaptive scenario 20th
percentile Mean 80th
percentile
Increased ramp construction and maintenance costs $0.20 $0.28 $0.36
Total: Adaptive scenario $39.51 $55.57 $69.75
Critical conditions scenario 20th
percentile Mean 80th
percentile
Modal shifting $2.72 $64.49 $111.81
Production loss $6.03 $149.15 $263.22
Total: Critical conditions scenario $9.10 $213.64 $377.64
Figures 4 and 5 show how this probability is distributed around the mean. As shown, the
distribution is fairly tightly packed around the mean; however, a long tail of probability above
the mean reflects uncertainty over the climate variables in the future. The probability that the
actual value is greater than the mean of $55 million is in the order of 60%, with a maximum
value of $155 million.
Figure 4: Adaptive scenario
For the critical conditions scenario (Figure 5), production loss dominates the expected total cost
of $213 million. Modal shifting (the increased operating costs to get materials to the mines) at
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
0%
2%
4%
6%
8%
10%
12%
14%
16%
18%
$14
$21
$28
$35
$42
$49
$56
$63
$70
$77
$84
$91
$98
$105
$112
$119
$126
$133
$140
$147
$154
$ millions
Mean is $55 million
YUKON RESEARCH CENTRE
Case Study of the Tibbitt to Contwoyto Winter Road
36
$65 million represents about 30% of the total cost. The simulation reveals a significant tail
above the mean, with 60% of the probability sitting above the mean and climbing to a
maximum value of $1.8 billion. Small changes in the assumed forecast of operational season
length have a significant impact on the results. This implies that the results are highly sensitive
to this assumption of operational season length, and that small changes in the season length
would result in large and significant future damages under the critical conditions scenario.
Figure 5: Critical conditions scenario
Of the climate variables, operational season length is the most important cost driver. This is
partially because some climate-cost variable interactions that are identified as key risks from a
cost perspective are likely to become less of a concern in the future. This is because the
associated climate risks become less severe in the climate forecast. Extreme cold, for instance,
is expected to become less pronounced over time. Other variables, such as the incidence of
temperatures swings in excess of 18°C, are not expected to change significantly, which means
they are not expected to drive any changes in baseline costs. The amount of snow on the
ground January 1 is expected to decrease over the forecast period, meaning that risks exist only
with respect to portage construction, and not for winter road construction. The incidence of
consecutive days of above-zero temperatures is expected to increase over the forecast period,
but not enough to trigger an adaptive scenario.
The lesser impact of some climate-cost variable interactions can also be explained by their
relatively smaller incremental costs. For example, carriers’ flexible scheduling costs are in the
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
0%
5%
10%
15%
20%
25%
30%
$-
$90
$180
$270
$360
$450
$540
$630
$720
$810
$900
$990
$1,080
$1,170
$1,260
$1,350
$1,440
$1,530
$1,620
$1,710
$1,800
$ millions
Mean is $213 million
FINAL REPORT
Economic Implications of Climate Change Adaptations for Mine Access Roads
in Northern Canada
37
aggregate much more significant than road operators’ construction costs, even when the
adaptive scenario is triggered for both.
The length of the operational season is by far the most significant climate variable because of its
significant impact on carrier costs in the adaptive scenario and its impacts on modal shifting and
production costs for mines in the critical conditions scenario. This is a significant concern for the
TCWR, since the length of the operational season is trending downward in the climate forecasts
(averaging 49 days by 2050) and has a significant degree of annual variability. The results
indicate that the expected evolution of the operational season length variable creates a
significant economic risk to road users in the future if demand continues at the levels seen
between 2002 and 2012. These costs would become even more pronounced if the demand for
road transport increases.
5.0 CONCLUSIONS
Although the Tibbitt to Contwoyto Winter Road is likely to support ongoing operations for the
mines that it now serves for a number of years, future climate variability and potential increases
in demand for road transport could lead to the recurrence of events similar to those in the 2006
season.
With the development of the Gahcho Kué mine and other potential future projects in the area,
road demand is likely to increase. The Gahcho Kué mine is planned to become operational in
2016, and therefore has a large number of shipments that need to be transported by winter
road in the 2015 season. To date they have managed to move at least 75% of those supplies and
will likely achieve the full 100% (Miller 2015). Miller (2015) notes that getting those shipments
through before the end of the road season is at least partly the result of good planning by the
company. Careful planning and flexible scheduling will help road users achieve the maximum
use possible; however, it is important for users to understand the costs incurred due to flexible
scheduling, and the costs that will be incurred if the season is shorter than the threshold for
adaptation.
Currently, the impact of expanded use due to the Gahcho Kué mine is manageable, but if more
projects are implemented there will be impacts on the ability of current mines to continue
flexible scheduling and transport the required number of loads per season. Any future decisions
around increasing the use of the road should take into account the possibility of shortened
seasons.
The Government of the Northwest Territories is exploring options to build a permanent all-
season road for the first half of the TCWR (Quenneville 2015a). This means that the most
vulnerable sections of the road — the southern sections and the portages — may no longer be
YUKON RESEARCH CENTRE
Case Study of the Tibbitt to Contwoyto Winter Road
38
problematic. Construction of a permanent all-season road could extend the season by a month,
or possibly longer (Quenneville 2015a). It could also significantly reduce the likelihood of
triggering the adaptation scenario or crossing into the critical conditions scenario during the life
of the current mine projects.
The TCWR is unusual compared to most other northern ice roads in that its main purpose is
solely to supply mining projects. Although having a longer season would improve access for the
TCWR, it is possible to move the required goods and equipment through flexible scheduling of
loads in the existing operational season.
Most northern ice roads also service northern communities, or supply projects that require
roads to be open as long as possible. In those cases, communities are concerned with having
consistent road access for as long as possible, as opposed to achieving a certain number of
loads. The lessons learned from the TCWR can inform those people making decisions for other
ice roads on how to manage a road to extend the season as long as possible. The economic
lessons from this report may not address the challenges faced by other ice roads, as flexible
scheduling is not an adaptation that can be applied to all circumstances. For winter ice roads in
areas currently at risk of warmer winters, building all-season sections in the most vulnerable
areas may be the most viable solution.
Although shorter and warmer winters can be expected as a result of climate change, seasonal
forecasting cannot yet predict with a high level of certainty when warmer winters will occur, or
when temperature swings will be extreme enough to affect road operations. Managing the road
with careful planning and preparation will allow road users to take advantage of flexible
scheduling by shipping as early as possible to avoid the risks of a shortened season.
FINAL REPORT
Economic Implications of Climate Change Adaptations for Mine Access Roads
in Northern Canada
39
REFERENCES
Ashbury, D. 2006. “Over the top.” Rio Tinto Review, No. 79: 9–14.
Babson College. 2011. Babson research: Economic importance of ice in the Arctic.
www.babson.edu/news-events/babson-news/pages/110628-babson-research-
economic-importance-of-ice-in-the-arctic.aspx (accessed March 28, 2014).
Comfort, G. and R. Abdelnour. 2013. Ice thickness prediction: A comparison of various practical
approaches. Prepared for CGU HS Committee on River Ice Processes and the
Environment, 17th Workshop on River Ice, Edmonton, Alberta, July 21–24, 2013, 1 p.
EBA. 2001. Tibbitt to Contwoyto Winter Road Environmental Setting report. EBA Engineering
Consultants LTD., Yellowknife, NT, 72 pp.
Engineers Canada. 2014. PIEVC Engineering Protocol – Version 10.
www.pievc.ca/e/doc_list.cfm?dsid=43 (accessed February 15, 2014).
Environment Canada. 2014. Historical Climate Data. Environment Canada, Ottawa, Ontario.
www.climate.weather.gc.ca/data_index_e.html.
Finlayson, D. 2007. “Ice road operators draw up trio of alternatives.” Edmonton Journal, July 4,
2007, www.canada.com/story.html?id=06770226-94e9-4570-8117-f802063f4cd8
(accessed December 12, 2013).
Fuss, S., J.G. Canadell, G.P. Peters, M. Tavoni, R.M. Andrew, P. Ciais, R.B. Jackson, C.D. Jones, F.
Kraxner, N. Nakicenovic, C. Le Quéré, M.R. Raupach, A. Sharifi, P. Smith and Y.
Yamagata. 2014. “Betting on negative emissions.” Nature Climate Change, Vol. 4: 850–
853. doi:10.1038/nclimate2392.
Galloway, J.M., A. Macumber, R.T. Patterson, H. Falck, T. Hadlari and E. Madsen. 2010.
Paleoclimatological assessment of the southern Northwest Territories and implications
for the long-term viability of the Tibbitt to Contwoyto Winter Road. Part I: Core
collection. Northwest Territories Geoscience Office, NWT Open Report 2010-002, 21 pp.
GNWT, GN and YG (Government of the Northwest Territories, Government of Nunavut and
Government of Yukon). 2011. Pan-Territorial Adaptation Strategy: Moving forward on
climate change adaptation in Canada’s North, 30 pp.
www.northernadaptation.ca/sites/default/files/Pan-
Territorial_Adaptation_Strategy.pdf.
Greenspan, L. 2008. “Diavik Diamond Mines, Inc.: Ice road conquerors.” Industry Today, Vol. 11,
Issue 2. http://industrytoday.com/article_view.asp?ArticleID=F296 (accessed December
12, 2013).
YUKON RESEARCH CENTRE
Case Study of the Tibbitt to Contwoyto Winter Road
40
Hayley, D. and S. Proskin. 2008. Managing the safety of ice covers used for transportation in an
environment of climate warming. In J. Locat, D. Perret, D. Turmel, D. Demers and S.
Leroueil (eds.). Proceedings of the 4th Canadian Conference on Geohazards: From causes
to management. Laval: Presse de l’Université Laval, 594 pp.
INAC (Department of Indian and Northern Affairs Canada). 2009. Canada’s Northern Strategy:
Our North, Our Heritage, Our Future, 41 pp.
http://publications.gc.ca/collections/collection_2009/ainc-inac/R3-72-2008.pdf.
IPCC (Intergovernmental Panel on Climate Change). 2013. Climate Change 2013: The Physical
Science Basis. Working Group I Contribution to the IPCC Fifth Assessment Report.
www.ipcc.ch/report/ar5/wg1.
Jarvis, K. and S. Proskin. 2011. “Across the ice.” Mining Magazine, November 2011: 38–42.
https://sensoft.ca/SenSoft/media/Resources/White-Papers/Across-the-Ice.pdf
(accessed March 5, 2014).
JVMC (Joint Venture Management Committee). 2014. Historical Winter Road Statistics (2002–
2012). www.jvtcwinterroad.ca/Facts (accessed February 5, 2015).
JVMC (Joint Venture Management Committee). 2013. Tibbitt to Contwoyto Winter Road Joint
Venture. www.jvtcwinterroad.ca/jvwr (accessed February 6, 2014).
JVMC (Joint Venture Management Committee). 2012. Winter road regulations and rules of the
road, December 2012, 27 pp.
McGregor, R.V., M. Hassan and D. Hayley. 2008. Climate Change Impacts and Adaptation: Case
studies of roads in northern Canada. Transportation Association of Canada, Toronto, 15
pp.
Mesher, D.E., S.A. Proskin and E. Madsen. 2008. Ice road assessment, modeling and
management. Prepared for Transportation Association of Canada Conference 2008,
Toronto, 20 pp.
Miller, J. 2015. Mountain Province expects to complete supply shipments to Gahcho Kué.
www.diamonds.net/News/NewsItem.aspx?ArticleID=51666&ArticleTitle=Mountain+Pro
vince+Expects+to+Complete+Supply+Shipments+to+Gahcho+Kue (accessed March 18,
2014).
Natural Resources Canada. 2013. Examining the long-term viability of an NWT ice road. Polar
Continental Shelf Program Science Report 2011/2012, pp. 32–34.
www.innovation.ca/sites/default/files/Rome2013/files/Polar%20Continental%20Shelf%
20Program,%20Science%20Report,2011-2012.pdf (accessed December 16, 2013).
FINAL REPORT
Economic Implications of Climate Change Adaptations for Mine Access Roads
in Northern Canada
41
Perrin, A., S. Eng, E. Sparling, H. Auld, N. Comer, A. Bevington, L. Kinnear. 2015. Enhancing
knowledge transfer to decision-makers with respect to climate change impacts on the
cryosphere. Northern Climate ExChange, Yukon Research Centre, Yukon College.
Nelson, J. and R. Schuchard. 2011. Adapting to climate change: A guide for the mining industry.
BSR Industry Series, 10 pp.
NOR-EX Ice Engineering. 2014. Ice Engineering, Optimization and Quality Assurance Program.
www.norexice.com/newwebsite/wp-content/uploads/2014/12/Diavik-
Diamond_Project-Profile.pdf (accessed December 15, 2014).
NRTEE (National Round Table on the Environment and the Economy). 2011. Paying the price:
The economic impacts of climate change for Canada, 162 pp.
www.fcm.ca/Documents/reports/PCP/paying_the_price_EN.pdf (accessed June 24,
2013).
Nuna Logistics. 2014. Tibbitt to Contwoyto Winter Road Joint Venture Ltd.
http://nunalogistics.com/clients/winter_road.html (accessed March 5, 2014).
Pearce, T., J.D. Ford, J.Prno, F. Duerden, L. Berrang-Ford, T. Smith, J. Pittman, A. Reid, M.
Beaumier and D. Marshall. 2009. Climate Change and Canadian mining: Opportunities
for adaptation. Prepared by ArcticNorth Consulting for the David Suzuki Foundation, 160
pp.
Prentice, B. 2013. “Ice roads, airships could work together.” Winnipeg Free Press, April 22, 2013.
www.winnipegfreepress.com/opinion/analysis/ice-roads-airships-could-work-together-
204044251.html (accessed February 5, 2015).
Quenneville, G. 2015a. All-season road to N.W.T. mines could extend ice road use by a month.
CBC News, January 30, 2015. www.cbc.ca/news/canada/north/all-season-road-to-n-w-t-
mines-could-extend-ice-road-use-by-a-month-1.2937351 (accessed February 5, 2015).
Quenneville, G. 2015b. Diamond mine ice road expected to have record year. CBC News,
February 1, 2015. www.cbc.ca/news/canada/north/diamond-mine-ice-road-expected-
to-have-record-year-1.2937270 (accessed February 5, 2015).
Rawlings, M., R. Bianchi and R. Douglas. 2009. Winter roads and ice bridges: Anomalies in their
seasonal usage and what we can learn from them. In: Economic Implications of Climate
Change. Annual Conference of the Transportation Association of Canada, Vancouver, 15
pp.
Rodan, G. 2012. “Storm strands truckers.” Northern News Services, March 19, 2012.
www.nnsl.com/preview/newspapers/stories/mar21_12rd.html (accessed April 15,
2014).
YUKON RESEARCH CENTRE
Case Study of the Tibbitt to Contwoyto Winter Road
42
Sawyer, D. 2014. The Economic Implications of Climate Change on Transportation Assets: An
analysis framework. London: International Institute for Sustainable Development.
www.iisd.org/publications/economic-implications-climate-change-transportation-
assets-analysis-framework.
Sherk, E. 2007. “Beyond the ice road.” Northern News Services, July 16, 2007.
www.nnsl.com/frames/newspapers/2007-07/jul16_07rd.html (accessed December 12,
2013).
Tetra Tech EBA. 2014. Ice road project wins NAPEGG award.
www.eba.ca/index.php/newsroom/ice-road-project-wins-napegg-award.html (accessed
March 11, 2014).
Tetra Tech EBA. 2013. Tibbitt to Contwoyto Winter Road, Northwest Territories and Nunavut.
www.tetratech.com/projects/tibbitt-to-contwoyto-winter-road-northwest-territories-
and-nunavut.html (accessed February 6, 2014).
FINAL REPORT
Economic Implications of Climate Change Adaptations for Mine Access Roads
in Northern Canada
43
APPENDIX A: HISTORICAL CLIMATE ANALYSIS
Tibbitt to Contwoyto Mining Road
Historical Climate Analysis
Prepared for:
Northern Climate ExChange (NCE)
Yukon Research Centre, Yukon College
March 31, 2014
YUKON RESEARCH CENTRE
Case Study of the Tibbitt to Contwoyto Winter Road
44
A. 1 INTRODUCTION
The Tibbitt to Contwoyto Winter Road (TCWR) is a private winter road that services mining
locations to the northeast of Yellowknife; it has been in seasonal service since 1982. Originally,
the road was 600 km in length; it now ends around the 400-km mark. It consists of both lake
crossings (87% of length) and portages (13%), which are intricately related to the yearly climate.
The road season is primarily during the months of February and March and averages 67 days
(JVMC 2013), although weather affects road construction and operations significantly from year
to year.
Appendix A investigates some of the historical climate indicators that affect road construction
and operation. The closest long-term climatological location is Yellowknife A (Yellowknife
airport), which has climate data dating to 1942. A second location, with interpolated conditions,
is located on Gordon Lake, northeast of Yellowknife. These locations, and the winter road route
to Contwoyto, are discussed in more detail in section A.2.1.
This appendix identifies climatological relationships on a year-to-year basis, and compares them
to operating days of the road for years for which data are available. Factors that show some
major influence on operating season length are discussed, along with those that show little
association. These factors are summarized in Table A13.
A.2 HISTORICAL CLIMATE FACTORS
Both temperature and precipitation are important to winter road construction. Temperature
affects freezing, while precipitation in the form of snow or rain can affect the road in both
positive and negative ways.
For most of the road length, data on these indicators are not available, so assumptions related
to conditions at Yellowknife are used to generate climatological relationships. These
relationships were then applied to interpolated conditions at Gordon Lake. The interpolated
conditions were extracted from a Canada-wide dataset called CANGRD, produced by Natural
Resources Canada and Environment Canada. The details of the CANGRD methodology for daily
data can be found in McKenney et al. 2011. The relationship between observations at
Yellowknife A and the Yellowknife CANGRD values were first evaluated to determine the
CANGRD dataset applicability. The agreement provides strength for the use of this dataset at
the interpolated location at Gordon Lake.
The climate factors considered for the winter road in this report are shown in Table A1. As
shown, these factors can have either positive or negative effects on the road. The primary
controller was expected to be freezing-degree day (FDD) accumulation, without which road
FINAL REPORT
Economic Implications of Climate Change Adaptations for Mine Access Roads
in Northern Canada
45
construction cannot begin. These factors were evaluated on a yearly basis from the Yellowknife
A daily observations. Road opening dates and season length information, where available, were
measured against these factors. It is important to note that the opening date of the winter road
is more likely controlled by climate than the closing date is. The road may be closed early if all
required materials have been transported, even if it is still useable.
Table A1: Climate factors related to the winter road
1 Winter temperature Colder — positive factor
2 Accumulated freezing-degree days (FDD) Higher accumulation — positive factor
3 Accumulated melting-degree days (MDD) Higher accumulation — negative factor
4 Date of accumulated FDD at 300 threshold Earlier date — positive factor
5 Accumulated snowfall at January 1
(the sum of observed daily snowfall)
Higher accumulation — negative for the ice/lakes
portion of the road, but positive for the land
portion of the road. Since ice formation is critical,
this is considered here as a predominantly
negative factor
6 Snow on ground at January 1
(measured snow on ground)
As above – predominantly a negative factor
7 Accumulated rainfall in November and
December prior to the winter road season
Higher accumulation — a negative factor, since it
contributes to ice/snow melt
8 Days with a 24-hour temperature change
greater than 18°C in November to April
Higher value — a negative factor, since it
contributes to the formation of ice road cracks
9 Days with a mean temperature above 0°C in
November to April
Higher value —a negative factor, since it
contributes to ice/snow melt
10 Days with a 24-hour temperature drop
greater than 20°C in November to April
Higher value — a negative factor, since it
contributes to the formation of ice road cracks
11 Observed ice thickness (Yellowknife, Great
Slave Lake)
Higher value — a positive factor
A.2.1 Winter Temperature
The most recent normal period (1981–2010) for Yellowknife shows an average winter
(December–February) temperature of –23.5°C. Winter temperatures are a prime factor in the
successful construction of winter roads, since both ground freezing and lake ice are required.
Using the standard 30-year normal period, previous values of winter average temperatures can
YUKON RESEARCH CENTRE
Case Study of the Tibbitt to Contwoyto Winter Road
46
be compared as an indication of historical winter temperature trends (Figure A1). Since the
earliest normal period, the average winter temperature has increased from –26°C to –23.5°C;
monthly changes are greatest during January and February, with an increase of up to 3.1°C
(Figure A2).
Figure A1: Historical average temperatures (°C) observed at Yellowknife A for five previous
normal periods
Data source: Environment Canada 2014
The observed trend of increasingly warmer winters presents a long-term challenge to winter
road operations. Although average temperatures are still very cold, increasingly warmer
average winter temperatures could affect the rate of natural ice growth and the effectiveness of
ice-thickening measures.
-30
-28
-26
-24
-22
-20
-18
-16
-14
-12
-10
1942-1970
1951-1980
1961-1990
1971-2000
1981-2010
Temperature (°C)
Dec
Jan
Feb
Mar
Winter(D-J-F)
FINAL REPORT
Economic Implications of Climate Change Adaptations for Mine Access Roads
in Northern Canada
47
Figure A2: Change in monthly temperature normals at Yellowknife A (1942–1970 to 1981–
2010)
Data source: Environment Canada 2014
Two maps of mean winter temperatures show the progression of warming temperatures
between 1961 and 1990 and the most recent normal period of 1981–2010 (Figures A3 and A4).
Progressively colder winter temperatures are encountered moving north in the TCWR region,
with a mean most recently of –24°C in the south at Yellowknife, and the mean temperature at
Contwoyto closer to –28°C. These maps are generated from the CANGRD dataset (McKenney et
al. 2011). Because the winter road is primarily affected by temperature, areas along the
southernmost portion of the route are more likely to be affected by warming temperatures. For
this reason, a point near the centre of Gordon Lake was also selected for investigation and is
discussed in section A.2.2.
0
0.5
1
1.5
2
2.5
3
3.5
Dec Jan Feb Mar
Temperature Change (°C)
+1.8
+3.1
+2.9
+1.9
YUKON RESEARCH CENTRE
Case Study of the Tibbitt to Contwoyto Winter Road
48
Figure A3: Mean winter (D-J-F) temperature along the TCWR, 1961–1990
Data source: McKenney et al. 2011
Figure A4: Mean winter (D-J-F) temperature along the TCWR, 1981–2010
Data source: McKenney et al. 2011
FINAL REPORT
Economic Implications of Climate Change Adaptations for Mine Access Roads
in Northern Canada
49
A.2.2 Freezing-Degree Days (FDD) and CANGRD temperature values
To provide a temperature record at Gordon Lake where none exists, the interpolated CANGRD
temperature dataset was used. It involves a carefully interpolated and gridded procedure
developed by McKenney et al. at 10-km resolution of daily maximum, minimum and mean
temperature and precipitation (2011). To ensure its validity in this region, the time series of
accumulated Freezing-Degree Days (FDD) using both datasets were calculated and compared
for Yellowknife A. FDD sums the daily mean temperature values falling below 0°C, so a day with
a mean temperature of –20°C would contribute 20 FDD. If the next day were –12°C, it would
contribute 12 FDD, leading to an accumulation of 32 FDD for the two days. Days above 0°C
contribute 0 FDD.
There was strong agreement between the observed temperatures (and therefore, FDD)
between both datasets. The accumulated FDD for both datasets for the period of 1961–2010 is
presented below for Yellowknife in Figure A5. FDD starts at zero on July 1 and is accumulated
over the winter season to June 30 of the following year.
Figure A5: Accumulated FDD, Yellowknife A and Yellowknife CANGRD, 1961–2009
Data source: Environment Canada 2014; McKenney et al. 2011
Typically, FDD accumulation begins near Julian day 300 (November) and reaches its maximum
near Julian day 100 (mid-April). The Environment Canada records for Yellowknife A show the
profiles for the years with the highest and lowest FDD accumulation (Figure A6). The CANGRD
profile would be indistinguishable from this observed profile calculated at Yellowknife A.
2,000
2,500
3,000
3,500
4,000
4,500
5,000
1961
1964
1967
1970
1973
1976
1979
1982
1985
1988
1991
1994
1997
2000
2003
2006
2009
Yellowknife A
Yellowknife Cangrd
YUKON RESEARCH CENTRE
Case Study of the Tibbitt to Contwoyto Winter Road
50
Figure A6: FDD accumulation profiles for warmest and coldest years at Yellowknife A
Data source: Environment Canada 2014
Figure A7: Correlation of FDD accumulation and mean winter temperature at Yellowknife A,
1942–2010
Data source: Environment Canada 2014
There is a high correlation between the accumulated FDD for each season and the mean winter
temperature. Colder years generate higher FDD values than warmer years do. In 2006, a warm
year, the FDD accumulation was close to 2,400; for the coldest year, 1967, accumulation was
0
500
1,000
1,500
2,000
2,500
3,000
3,500
4,000
4,500
182
198
214
230
246
262
278
294
310
326
342
358
9
25
41
57
73
89
105
121
137
153
169
Julian Day
1967-cold year
2006-warm year
November
mid-April
Freezing
season
y = -150.18x - 199.08
R² = 0.9932
3,300
3,350
3,400
3,450
3,500
3,550
3,600
3,650
3,700
3,750
-26.5 -26 -25.5 -25 -24.5 -24 -23.5 -23
FDD Accumulation
Mean Winter Temperature of the Normals Period (°C)
1942-1970
1951–1980
1961–1990
1971–2000
1981–2010
FINAL REPORT
Economic Implications of Climate Change Adaptations for Mine Access Roads
in Northern Canada
51
almost 4,000. Figure A7 shows the correlation between the average accumulated FDD for each
of the previous normal periods and the mean winter temperature.
As might be expected, with the increase in mean winter temperatures, FDD accumulation has
correspondingly decreased, from a value of 3,700 to approximately 3,200, as seen in Figure A8.
There is high year-to-year variation: 1967 was the best year for ice formation and 2006 was the
worst. The average yearly FDD accumulation (shown by the black line) is 3,521 over the period.
Figure A8: FDD accumulation at June 30, Yellowknife A, 1943–2013
Data source: Environment Canada 2014
To determine good versus bad years for this indicator, the mean value is used: good years
(green) have higher-than-mean values and bad years (pink) have lower-than-mean values. This
can be seen in Table A2 for the 1943–2013 period. There is a trend of increasingly bad years due
to warming winters, which reduces the accumulated FDD for the season.
Table A2: FDD accumulation, 1943–2013
Data source: Environment Canada 2014
A.2.2.1 FDD and winter road operating period
With the data on the length of the winter road operational season since 1994 (JVMC 2013), the
relationship between FDD accumulation and operational season length can be investigated. The
correlation is not ideal, since other factors besides FDD affect the road season length. Some of
2,000
2,500
3,000
3,500
4,000
4,500
5,000
1943
1946
1949
1952
1955
1958
1961
1964
1967
1970
1973
1976
1979
1982
1985
1988
1991
1994
1997
2000
2003
2006
2009
2012
YUKON RESEARCH CENTRE
Case Study of the Tibbitt to Contwoyto Winter Road
52
these factors are not related to climate, such as the mining economy of that year, or whether all
the materials necessary have already been transported for a particular year. Nevertheless, there
is a broad association between the two variables, as seen in Figure A9, with a greater number of
operating days associated with higher FDD accumulation. Simply put, colder years lead to more
operating days.
Figure A9: Comparison of TCWR operating days and FDD accumulation, 1994–2013
Data source: Environment Canada 2014; JVMC 2013
A.2.2.2 FDD and estimated ice thickness
Relationships also theoretically exist between accumulated FDD and the generation of ice
(CRREL 2004). A higher accumulation of FDD increases possible ice thickness. It follows the
following formula: T = C x FDDacc0.5 where T=ice thickness inches; C= a calibration factor related
to waves, exposure, and snow; and FDDacc = accumulated Freezing-Degree Days in °F.
The C factor typically ranges between 0.2 and 0.8 during ice formation; once ice is formed this
factor is ignored. These calculations used this formula with the maximum FDD accumulation at
the end of the season, so C was not considered. This provided a theoretical ice thickness based
on the FDD accumulation. The observations here were adjusted for use in this equation (which
is based on imperial units) to produce a metric value for ice thickness. Applying the maximum
yearly FDD accumulation to the equation, the resulting yearly theoretical ice thickness was
calculated, ignoring all other factors (for example, snow cover or waves on lakes, which would
reduce thickness). The calculation resulted in an average theoretical ice thickness of 211 cm at
the beginning of the record, trending lower to an average ice thickness of 193 cm in recent
y = 0.0185x + 3.6313
R² = 0.2909
30
40
50
60
70
80
90
100
2,000 2,200 2,400 2,600 2,800 3,000 3,200 3,400 3,600 3,800 4,000
Road Operating Days
Accumulated FDD (July 1 to June 30)
FINAL REPORT
Economic Implications of Climate Change Adaptations for Mine Access Roads
in Northern Canada
53
years, a reduction of 18 cm (Figure A10). An accumulation of 300 FDD corresponds roughly to an
ice thickness of 25 cm.
Figure A10: Estimated Ice thickness (cm) based on FDD accumulation using C coefficient at 1.0,
1943–2013
Data source: Environment Canada 2014
Environment Canada, through the Canadian Ice Service, has a record of ice thickness
measurements at Yellowknife (taken nearby on Great Slave Lake) (CCIN 2013). The maximum
ice thickness measurement for each season was extracted from this data and plotted in Figure
A11 (there is a gap in the data between 1997 and 2004).
Figure A11: Observed annual maximum ice thickness, Yellowknife, 1959–2013
Data source: CCIN 2013
150
160
170
180
190
200
210
220
230
1943
1946
1949
1952
1955
1958
1961
1964
1967
1970
1973
1976
1979
1982
1985
1988
1991
1994
1997
2000
2003
2006
2009
2012
80
100
120
140
160
180
200
1959
1961
1963
1965
1967
1969
1971
1973
1975
1977
1979
1981
1983
1985
1987
1989
1991
1993
1995
1997
1999
2001
2003
2005
2007
2009
2011
2013
Ice Thickness (cm)
YUKON RESEARCH CENTRE
Case Study of the Tibbitt to Contwoyto Winter Road
54
The observed maximum thickness is always less than the theoretical ice thickness that was
calculated based on FDD accumulation. This is to be expected, since negative factors that
influence ice growth are not considered in the theoretical calculation. Although the FDD-
calculated ice thickness for 2006 is at a record low, this does not show up in the observed
record at Yellowknife. There is an association between the two datasets (both decreasing over
time), so the calculation is a meaningful indicator, although weak (Figure A12).
Figure A12: Comparison of observed and theoretical maximum ice thickness at Yellowknife
Data source: CCIN 2013
A.2.3 Melting-Degree Days (MDD)
This variable serves as an indicator of the potential closing of the winter road due to the
accumulation of heat from days with a mean temperature greater than 0°C. MDD is the
opposite of FDD. The summation starts on January 1 of each year; the value reached as of April
30 (which is beyond the winter road typical closure date) is plotted in Figure A13. The higher
accumulation of MDD during the January-to-April period would be a negative influence on
winter road operations because of its contribution to melting conditions.
160